Cost-Utility of First-Line Disease-Modifying Treatments for Relapsing-Remitting Multiple Sclerosis
Full text
Clinical The apeu ics/Volume 39, Numbe 3, 2017
Cos -u ili y o Fi s -line Disease-modi ying
T ea men s o Relapsing–Remi ing
Mul iple Scle osis
E kki Soini, MSc
1
; Jaana Jou seno, MSc
2
; and Ma ja-Liisa Sumelah i, MD
3
1
ESiOR Oy, Kuopio, Finland;
2
Genzyme (a Sano i Company), Helsinki, Finland; and
3
School o Medicine,
Uni e si y o Tampe e, Tampe e, Finland
ABSTRACT
Pu pose: This s udy e alua ed he cos -e ec i eness
o fi s -line ea men s o elapsing– emi ing mul iple
scle osis (RRMS) (dime hyl uma a e [DMF] 240 mg
PO BID, e iflunomide 14 mg once daily, gla i ame
ace a e 20 mg SC once daily, in e e on [IFN]-β1a
44 mg TIW, IFN-β1b 250 mg EOD, and IFN-β1a 30 mg
IM QW) and bes suppo i e ca e (BSC) in he heal h
ca e paye se ing in Finland.
Me hods: The p ima y ou come was he modeled
inc emen al cos -e ec i eness a io (ICER; €/quali y-
adjus ed li e-yea [QALY] gained, 3%/y discoun ing).
Ma ko coho modeling wi h a 15-yea ime ho izon
was employed. Du ing each 1-yea modeling cycle,
pa ien s ei he main ained he Expanded Disabili y S a us
Scale (EDSS) sco e o expe ienced p og ession, de eloped
seconda y p og essi e MS (SPMS) o showed EDSS
p og ession in SPMS, expe ienced elapse wi h/wi hou
hospi aliza ion, expe ienced an ad e se e en (AE), o
died. Pa ien s' cha ac e is ics, RRMS p og ession p oba-
bili ies, and s anda dized mo ali y a ios we e de i ed
om a egis y o pa ien s wi h MS in Finland. A mixed-
ea men compa ison (MTC) in o med he ea men
e ec s. Finnish Eu oQol Fi e-Dimensional Ques ionnai e,
Th ee-Le el Ve sion quali y-o -li e and di ec -cos
es ima es associa ed wi h EDSS sco es, elapses, and
AEs we e applied. Fou app oaches we e used o assess
he ou comes: cos -e ec i eness plane and e ficiency
on ie s ( ela i e alue o e ficien ea men s); cos -
e ec i eness accep abili y on ie , which demons a ed
op imal ea men o maximize ne benefi ; Bayesian
ea men anking (BTR); and an impac in es men
assessmen (IIA; a cos -benefi assessmen ), which
inc eased he clinical in e p e a ion and appeal o
modeled ou comes in e ms o absolu e benefi gained
wi h fixed d ug- ela ed budge . Robus ness o esul s
was es ed ex ensi ely wi h sensi i i y analyses.
Findings: Based on he modeled esul s, e ifluno-
mide was less cos ly, wi h g ea e QALYs, e sus
gla i ame ace a e and he IFNs. Te iflunomide had
he lowes ICER (24,081) e sus BSC. DMF b ough
ma ginally mo e QALYs (0.089) han did e ifluno-
mide, wi h g ea e cos s o e he 15 yea s. The ICER
o DMF e sus e iflunomide was 75,431. Te ifluno-
mide had 450% cos -e ec i eness p obabili ies wi h
a willingness- o-pay h eshold o o€77,416/QALY
gained. Acco ding o BTR, e iflunomide was fi s -bes
among he disease-modi ying he apies, wi h po en ial
willingness- o-pay h esholds o up o €68,000/QALY
gained. In he IIA, e iflunomide was associa ed wi h
he longes inc emen al quali y-adjus ed su i al and
ime wi hou cane use. Gene ally, p ima y ou comes
esul s we e obus , based on he sensi i i y analyses.
The esul s we e sensi i e only o la ge changes in
analysis pe spec i e o mixed- ea men compa ison.
Implica ions: The esul s we e sensi i e only o
la ge changes in analysis pe spec i e o MTC. Based
on he analyses, e iflunomide was cos -e ec i e
e sus BSC o DMF wi h he common h eshold
alues, was dominan e sus o he fi s -line RRMS
ea men s, and p o ided he g ea es impac on
in es men . Te iflunomide is po en ially he mos
cos -e ec i e op ion among fi s -line ea men s o
*
Selec ed da a om his a icle we e p esen ed in pos e o ma a he
31s Cong ess o he Eu opean Commi ee o T ea men and Resea ch
in Mul iple Scle osis, Ba celona, Spain, Oc obe 7–10, 2015; and in
pos e o ma a he 18 h Annual Eu opean Cong ess o he In e na-
ional Socie y o Pha macoeconomics and Ou comes Resea ch,
Milan, I aly, No embe 7–11, 2015 (Value Heal h 2015;18:A756).
Accep ed o publica ion Janua y 18, 2017.
h p://dx.doi.o g/10.1016/j.clin he a.2017.01.028
0149-2918/$ - see on ma e
&2017 The Au ho s. Published by Else ie HS Jou nals, Inc. This is an
open access a icle unde he CC BY-NC-ND license
(h p://c ea i ecommons.o g/licenses/by-nc-nd/4.0/).
Ma ch 2017 537
RRMS in Finland. (Clin The . 2017;39:537–557)
&2017 The Au ho s. Published by Else ie HS
Jou nals, Inc.
Key wo ds: cos -e ec i eness, dime hyl uma a e,
economic e alua ion, gla i ame ace a e, in e e on-β,
e iflunomide.
INTRODUCTION
Mul iple scle osis (MS)—a ch onic p og essi e, au o-
immune, inflamma o y disease—a ec s 42 million
people wo ldwide. App oxima ely 89% o cases a e
classified as elapsing– emi ing MS (RRMS) a he
ime o diagnosis.
1
MS p e alence is pa icula ly high
in he Uni ed Kingdom, he Uni ed S a es, Canada,
Ge many, and Scandina ia.
2,3
In Finland, MS p e a-
lence a ies egionally, om 100 o 200 pe 100,000
inhabi an s.
4–7
In young adul s wi h MS, p ognosis is based on
an indi idual’s ac o s.
1
The p og ession and
accumula ing disabili y cause a significan human
and economic bu den
8–15
and he need o suppo .
16
The isk o dea h among Finnish pa ien s wi h MS
is 2.8- old compa ed wi h ha in he gene al popula-
ion, being 3.4- old in women and 2.2- old in men as
ea ly as 2 o 10 yea s a e diagnosis.
17
Relapse, MS
p og ession, and disabili y le el (eg, highe Expanded
Disabili y S a us Scale [EDSS] sco e
18
) a e associa ed
wi h a highe isk o mo ali y,
17,19,20
addi ional
cos s,
9–14
and quali y o li e (QoL) losses.
9,10,12,14,21–24
MS ea men wi h disease-modi ying he apies
(DMTs) is aimed a dec easing he inflamma o y
ac i i y leading o elapses, s opping o slowing
p og ession o esidual disabili y, and, e en ually,
delaying he p og ession o he seconda y p og essi e
phase. Howe e , long- e m p ognosis among ea ed
pa ien s is la gely unknown. Based on Finnish d ug
eimbu semen and sales da a,
25
commonly used fi s -
line DMTs include injec able DMTs, namely gla i -
ame ace a e (GA), in e e on (IFN)-β1a IM, IFN-β1a
SC, and IFN-β1b SC.
Dime hyl uma a e (DMF) and e iflunomide a e
new o al DMTs eimbu sed as he fi s -line ea men
o RRMS in Finland. The e ficacy and sa e y o
DMF 240 mg BID o es ablished MS ha e been
s udied in he Phase III CONFIRM (Compa a o and
an O al Fuma a e in Relapsing-Remi ing Mul iple
Scle osis)
26,27
and DEFINE (De e mina ion o he
E ficacy and Sa e y o O al Fuma a e in Relapsing–
Remi ing MS)
28,29
ials (ClinicalT ials.go iden i-
fie s: NCT00451451 and NCT00420212, espec-
i ely). The e ficacy and sa e y o e iflunomide 14
mg once daily o es ablished MS ha e been demon-
s a ed in he Phase III TEMSO (Te iflunomide Mul i-
ple Scle osis O al Te iflunomide o Relapsing
Mul iple Scle osis)
30–33
and TOWER (Te iflunomide
O al in People Wi h Relapsing Mul iple Scle osis)
34,35
ials (NCT00134563 and NCT00751881, espec-
i ely), and in pa ien s wi h a fi s clinical episode
sugges i e o MS in he TOPIC (O al Te iflunomide
o Pa ien s wi h a Fi s Clinical Episode Sugges i e
o Mul iple Scle osis)
36
ial (NCT00622700).
E ec i eness o e iflunomide compa ed wi h
IFN-β1b SC has been demons a ed in he Phase III
TENERE (Te iflunomide and Rebi ® in Pa ien s
wi h Relapsing Mul iple Scle osis)
37
ial
(NCT00883337).
We e alua ed he cos -u ili y o injec able and o al
fi s -line DMTs in he Finnish popula ion o pa ien s
wi h RRMS, based on a decision-analy ical model. To
ou knowledge, he e a e no p e iously published
jou nal a icles on he cos -u ili y o fi s -line o al
DMTs in a Eu opean se ing o on o al and injec able
DMTs o fi s -line ea men o RRMS. In addi ion,
p og ession o RRMS in Finnish pa ien s has no been
assessed be o e, and he 4 di e en app oaches elab-
o a ing he key esul s om MS cos -u ili y analysis
ha e no been p e iously epo ed.
MATERIALS AND METHODS
The cos -u ili y o he fi s -line DMTs in he Finnish
RRMS popula ion was assessed in a decision-
analy ical modeling amewo k
38
by implemen ing
a Ma ko coho model wi h mu ually exclusi e
heal h s a es in Excel 2007, including Visual
Basic o Applica ions (Mic oso Co po a ion,
Redmond, Washing on). The modeling app oach
ollowed he Finnish guidance o heal h economic
analyses.
39
The p ima y ou come o analysis was he modeled
inc emen al cos -e ec i eness a io (ICER), epo ed
as Eu os pe quali y-adjus ed li e-yea (€/QALY)
gained. The in e p e a ion o ICER is challenging in
Finland because he decision make ’s willingness- o-
pay (WTP) h eshold pe QALY gained has no been
publicly decla ed,
40
and significan a ia ion in
Clinical The apeu ics
538 Volume 39 Numbe 3
decision make WTP be ween diseases may exis .
41
Based on ou expe ience, he UK h esholds
42,43
could be applicable in Finland, so ha alues o
o€25,000 o €25,000–37,000/QALY gained
would indica e mos plausible o plausible cos -e ec-
i eness, espec i ely; and, on a e age, €55,000/
QALY gained could be accep able o end-o -li e
ea men based on he UK popula ion-weigh ed
decisions. This applicabili y o UK h esholds is based
on he obse a ion ha many a icles om Fin-
land
41,44–55
ha e e e ed o a WTP h eshold o
€50,000/QALY gained, which is p obably based
on he so-called "dialysis a gumen ."
41
The Finnish
Medicines Agency has conside ed ha €68,000/QALY
gained app oaches he maximum cos -e ec i eness
h eshold o a li e- h ea ening cance
56
—a esul
well in line wi h ea lie Finnish a e age findings.
41
The heal h ca e paye se ing, which is ecom-
mended in he Finnish guidance o heal h economic
analyses,
39
was used in he modeling. This model
includes di ec heal h and social ca e cos s, and
excludes income ans e s ( axes) and indi ec cos s
(eg, ime cos s, disabili y paymen s, p esen eeism,
absen eeism, and in o mal ca e). A scena io analysis,
including p oduc i i y losses,
14
was pe o med o
assess he obus ness o his di ec -cos ing pe spec i e.
A summa y o he modeled key esea ch ques ions is
gi en in Table I as an ex ended PICO amewo k,
which is used o cap u e and cla i y he essen ial pa s
o complica ed cos -e ec i eness assessmen in a
sensible o de (namely, PICOSTEPS: P, pa ien s; I,
in e en ions; C, compa a o ; O, ou comes; S, se ing;
T, ime ho izon; E, e ec s; P, pe spec i e; and S,
sensi i i y analyses).
A ela i ely s aigh o wa d, limi ed cos –benefi
analysis (clinical alue analysis) app oach was ecen ly
de eloped.
46
As a seconda y complemen a y analysis,
an impac in es men assessmen (IIA) was ca ied ou
o inc ease he clinical appeal and in e p e a ion o he
p ima y ou come esul s.
46
The IIA he e co e ed a
fixed d ug- ela ed budge based on he mos a o d-
able DMT and inc emen al quali y-adjus ed su i al
o ime o cane use (EDSS sco e, 6) e sus bes
suppo i e ca e (BSC; ial compa a o ). The ou come
(impac on in es men [II]) o he IIA was he du a ion
o benefi ob ained in compa ison wi h BSC wi h he
fixed budge . This IIA inco po a ed an explici mini-
mal willingness- o-in es (WTI) alue o DMT based
on he mos a o dable DMT and, hus, demons a ed
he mean absolu e cos –benefi in e ms o a single
uni :
II¼D ug heal h bene i i s BSCðÞ
Assumed d ug ela ed minimal WTIðÞ
D ug ela ed cos i
ðÞ
(Equa ion 1)
whe e iindica es a pa icula d ug ea men .
Consequen ly, he esul o he IIA is a s anda dized
benefi (II) ob ained wi h he gi en WTI (in ac , he
WTI can be g ea e han he minimum assumed he e,
and he benefi inc eases acco dingly).
Pa ien s
Finland’s MS esea ch egis y da a we e used o
define he coho cha ac e is ics in he model. Based
on he MS esea ch egis y da a (713 ambula o y
pa ien s om Finland, wi h MS diagnosed in 1991–
2010 and an EDSS sco e o 0–6.5 obse ed a base-
line; see Supplemen al Ma e ial A in he online e sion
a h p://dx.doi.o g/10.1016/j.clin he a.2017.01.028),
he mean age o modeled pa ien s was 35.64 yea s,
and he emale/male a io was 2.57. The dis ibu ion
o EDSS sco es a baseline is shown in Figu e 1.
Model
The clinical cou se o MS was modeled
(Figu e 2)
73,74
o cap u e all ele an e idence,
38,39,43
as no di ec compa ison is cu en ly a ailable. Models
a e always hypo he ical and con ain an elemen o
unce ain y, bu when elying on conse a i e and ai
s uc u e and es ima es—and keeping he modeling
assump ions in mind— hey can p oduce use ul in o -
ma ion o decision making.
In he model shown in Figu e 2,pa ien s wi h
RRMS ei he main ained he same EDSS o ansi ed
o ano he EDSS heal h s a e as he disease p og essed,
de eloped seconda y p og essi e MS (SPMS),
ansi ed o ano he EDSS s a e in SPMS, o died
(EDSS sco e, 10; abso bing s a e) wi hin he 1-yea
model cycles. Wi hin each cycle, pa ien s expe ienced
a elapse (wi h/wi hou hospi aliza ion) and/o an
ad e se e en (AE). The ela i e e ec s o DMTs we e
implemen ed as modifie s o he modeled clinical
cou se o MS. Midcycle es ima es (li e- able me hod
o hal -cycle co ec ion
75–77
) we e used o a oid o e -
o unde es ima ion o modeled ou comes.
E. Soini e al.
Ma ch 2017 539
Disease P og ession
Disease p og ession and elapses we e modeled
independen ly. Disease p og ession in e ms o he
EDSS sco e de elopmen du ing RRMS was es ima ed
om Finland’s MS esea ch egis y da a, consis ing
o 2299 EDSS measu emen s. The p obabili y o
ansi ing om RRMS o SPMS was es ima ed, and
EDSS de elopmen du ing SPMS was based on esul s
Table I. PICOSTEPS: Summa y o he esea ch ques ions.
PICOSTEPS Desc ip ion
P: Pa ien s Finnish adul s wi h inciden RRMS and EDSS sco es 0.0–6.5 a baseline based on da a om
a Finnish MS egis y
I: In e en ions DMTs: DMF 240 mg PO BID, e iflunomide 14 mg once daily, GA 20 mg SC once daily, IFN-
β1a 44 mg SC TIW, IFN-β1b 250 mg SC EOD, IFN-β1a 30 mgIMQW
C: Compa a o Common compa a o : BSC ( ial placebo)
O: Ou comes P ima y: ICER gi en as he cos /QALY gained based on he di ec cos
Seconda y: disagg ega ed and o al QALYs (based on EQ-5D-3L) and cos s, li e-yea s, yea s
wi hou impai ed mobili y (EDSS o6; ie, yea s wi hou cane use), cos -e ec i eness plane
and e ficiency on ie s, cos -e ec i eness accep abili y on ie s, Bayesian ea men
anking, and cos –benefi assessmen . Discoun ing: 3%/y
S: Se ing P obabilis ic decision analy ical modeling (Ma ko coho model), including 21 heal h s a es
eflec ing he disease p og ession (modified by ea men e ficacy); and e en s eflec ing
elapses, AEs, and wi hd awals
T: Time ho izon 15 yea s, based on he ollow-up da a om he Finnish egis y, ime since diagnosis in a
Finnish cos and EQ-5D-3L MS s udy,
14
yea s co e ed by he B i ish Columbia, Canada,
egis y,
57,58
and app oxima e ime om RRMS o SPMS in he London On a io MS
egis y da abase. Fo he London On a io MS egis y o igins, see Weinshenke e al.
59
E: E ec s RRMS p og ession: Finnish MS egis y da a (see Supplemen al Ma e ial A in he online
e sion a h p://dx.doi.o g/10.1016/j.clin he a.2017.01.028). SPMS p og ession: London
On a io MS egis y (see Supplemen al Ma e ial A in he online e sion a h p://dx.doi.
o g/10.1016/j.clin he a.2017.01.028). Relapse a es: published elsewhe e.
21,60
Relapse-
associa ed hospi aliza ions: published elsewhe e.
30,32,33
Mo ali y: Finnish MS egis y
da a and s a is ics
61
wi h EDSS- ela ed
17
adjus men . EDSS-associa ed cos s and quali y
o li e: es ima ed om a Finnish s udy.
14
Relapse cos s: Finnish MS egis y da a. Relapse
disu ili y: Finnish s udy
14
accoun ing o hospi aliza ion s a us and du a ion.
23,24
12-wk esponses wi h DMT, annual elapse a es, and wi hd awals: mixed- ea men
compa ison.
62,63
DMT e ec s on elapses esul ing in hospi aliza ions: published
elsewhe e.
32,64,65
DMT cos s: d ugs,
66
moni o ing.
67–71
AEs: disu ili y,
72
du a ion, cos s,
and occu ence (see Supplemen al Ma e ial B in he online e sion a h p://dx.doi.o g/
10.1016/j.clin he a.2017.01.028).
P: Pe spec i e Finnish paye pe spec i e. A scena io analysis wi h a socie al pe spec i e.
S: Sensi i i y
analyses
25 de e minis ic scena ios: impac o modeling assump ions, esul obus ness, and
gene alizabili y
P obabilis ic sensi i i y analysis: join unce ain y o he inpu es ima es
AE ¼ad e se e en ; BSC ¼bes suppo i e ca e; DMF ¼dime hyl uma a e; DMT ¼disease-modi ying he apy; EDSS ¼
Expanded Disabili y S a us Scale; EQ-5D-3L ¼Eu oQol Fi e-Dimensional Ques ionnai e, Th ee-Le el Ve sion; GA ¼
gla i ame ace a e; ICER ¼inc emen al cos -e ec i e a io; IFN ¼in e e on; MS ¼mul iple scle osis; QALY ¼quali y-
adjus ed li e-yea ; RRMS ¼ elapsing– emi ing mul iple scle osis; SPMS ¼seconda y p og essi e mul iple scle osis.
Clinical The apeu ics
540 Volume 39 Numbe 3
om he London On a io egis y o MS (see
Supplemen al Ma e ial A in he online e sion a h p://
dx.doi.o g/10.1016/j.clin he a.2017.01.028). Fo he
o igins o egis y, see Weinshenke e al.
59
The elapse
a es in pa ien s no ecei ing DMTs we e aken om
published e e ences.
21,60
The pe cen age o elapses
leading o hospi aliza ion (30.7%) was es ima ed om
he TEMSO ial.
30,32,33
The annual p obabili y o dea h was modeled
based on Finland’s gene al popula ion mo ali y a es
by applying he obse ed MS emale/male a io o
2.57 om Finland’s MS esea ch egis y da a o
Finland’s all-cause age- and sex-specific mo ali y
a es om he yea 2014,
61
mul iplying he sex-
weigh ed gene al popula ion mo ali y a e by he
EDSS-specific s anda dized mo ali y a io, and con-
e ing he esul o gi e he p obabili y.
78
The EDSS-
specific s anda dized mo ali y a io was es ima ed
om Finland’s MS esea ch egis y esul s
17
by using
linea in e pola ion:
S anda dized mo ali y a io
¼0:515 EDSSþ1:000 (Equa ion 2)
T ea men E icacy and Tole abili y
T ea men e ficacy was assessed by common MS
s udy ou comes: sus aining he same disabili y s a us
o 12 weeks, annualized elapse a e (ARR), and
elapses. Pe sis ence was assessed by wi hd awal a es,
and ole abili y, by AEs. Rela i e a es o hospi al-
iza ion in he model we e de i ed om he ollowing
clinical ials: IFN-β1a SC, CARE MS I (Compa ison
o Alem uzumab and Rebi E ficacy in Mul iple
Scle osis)
64
(assumed o apply o GA and IFN-β1b
SC); IFN-β1a IM, TRANSFORMS (T ial Assessing
Injec able In e e on e sus FTY720 O al in
Relapsing–Remi ing Mul iple Scle osis)
65
; and
e iflunomide, TEMSO
32
(assumed o apply o
DMF). Wi hd awals we e assumed o happen a he
ini ia ion o a new model cycle (bu no a he s a
o he fi s cycle), and pa ien s we e assumed o
discon inue hei cu en ea men when hey
p og essed om RRMS o SPMS.
Disabili y p og ession, ARR, and wi hd awal a es
we e modeled based on a mixed- ea men compa -
ison assessed by he Na ional Ins i u e o Heal h
and Ca e Excellence.
62,63
To accoun o new MS
diagnos ics, ea lie ea men , and e idence o de-
c eased ARR o e ime, he base case analysis
included ials ha en olled Z80% o pa ien s who
had RRMS and had been ec ui ing pa ien s since
2000. In addi ion, mul iway sensi i i y analyses
(disabili y p og ession, ARR, wi hd awal a es) o
mixed- ea men compa ison wi hou yea limi and
wi h o wi hou adjus men o placebo elapses we e
pe o med.
T ea men sa e y was modeled using epo ed AEs
om clinical ials o ea lie heal h echnology assess-
men s, hei cos s, and QoL e ec s (see Supplemen al
Ma e ial B in he online e sion a h p://dx.doi.o g/
10.1016/j.clin he a.2017.01.028). AEs epo ed wi h
35
30
25
20
15
10
5
0EDSS 0 EDSS 1 EDSS 2 EDSS 3 EDSS 4 EDSS 5 EDSS 6
P opo ion o Pa ien s, %
26.79
33.10
12.06
5.47
2.95 3.51
16.13
Figu e 1. Expanded Disabili y S a us Scale
(EDSS) sco e dis ibu ion a he in-
i ia ion o modeling.
ansi ions may happen be ween EDSS 0-9 and o dea h
SPMS
Dea h
0123456789
0123456789
RRMS ansi ions may happen be ween EDSS 0-9, o SPMS and o dea h
Figu e 2. Simpli ied p esen a ion o he Ma ko
model and i s key heal h s a es. Re-
lapses and ad e se e en s a e no
depic ed. EDSS ¼Expanded Disabili y
S a us Scale; RRMS ¼ elapsing– emi -
ing mul iple scle osis; SPMS ¼sec-
onda y-p og essi e mul iple scle osis.
E. Soini e al.
Ma ch 2017 541
simila e ms we e assumed o be ea ed simila ly and
o esul in simila QoL loss.
Quali y-adjus ed Su i al
The Eu oQol Fi e-Dimensional Ques ionnai e,
Th ee-Le el Ve sion (EQ-5D-3L) QoL o EDSS
sco es was modeled on he basis o da a om
DEFENSE (Bu den o Illness in Mul iple Scle o-
sis),
14
a ecen c oss-sec ional su ey om Finland.
The occu ence and impac
72
o AEs (see
Supplemen al Ma e ial B in he online e sion a
h p://dx.doi.o g/10.1016/j.clin he a.2017.01.028)
and elapses
14,24
we e accoun ed o . Finland’s
EDSS- ela ed QoL alues
14
we e deemed
accep able because he mean EQ-5D-3L sco e in
EDSS 0-1 was in line wi h alues om he gene al
popula ion o Finland.
79
Howe e , he s udy om
Finland
14
did no speci y QoL ela ed o elapse
wi h and wi hou hospi aliza ions.
Findings om s udies sugges g ea e disu ili y o
elapse wi h hospi aliza ion compa ed wi h elapses
wi hou hospi aliza ion.
23,24
In a US s udy, he QoL
losses in elapsed pa ien s wi h and wi hou hospi al-
iza ion we e epo ed as –0.302 and –0.091, espec-
i ely.
24
The la e es ima e is simila o he Finnish
elapse loss, ha is, –0.064,
14
which used an ex ensi e
1-yea ecall pe iod and did no make a dis inc ion
be ween hospi alized and nonhospi alized pa ien s o
numbe o elapses.
To app oxima e he QoL loss associa ed wi h
hospi aliza ions, he Finnish QoL loss was weigh ed
wi h he obse ed a io be ween he QoL losses o
hospi alized and nonhospi alized elapses in he US
s udy
24
( a io –0.302/–0.091 ¼3.3187) o ob ain
disu ili y o hospi alized pa ien s in Finland. The
applied QoL losses in elapsed pa ien s wi h and
wi hou hospi aliza ion in he model we e –0.212
and –0.064, espec i ely. The QoL e ec o elapse
was assumed o las o 3 mon hs.
23
Cos s
Annual DMT cos was calcula ed using he indi-
ca ed mean dose o each d ug and numbe o doses
pe yea (365.25 d/y), de e mined o each ea men
egimen based on he p oduc labeling. Fo d ugs wi h
mul iple package sizes, he d ug cos s we e es ima ed
by weigh ing o he package cos s by hei es ima ed
ma ke sha e (Table II). A 100% dose in ensi y and
adhe ence we e assumed.
Adminis a ion, moni o ing (Table III), and AE
cos s (see Supplemen al Ma e ial B in he online
e sion a h p://dx.doi.o g/10.1016/j.clin he a.2017.
01.028) we e calcula ed on he basis o esou ce
consump ion mul iplied by he associa ed uni cos s.
DMT-associa ed esou ces we e based on he p oduc
labeling, ecommenda ions in Finland,
1,80,81
publica-
ions o ea lie assessmen s (see Supplemen al Ma e ial B
in he online e sion a h p://dx.doi.o g/10.1016/
j.clin he a.2017.01.028), and clinical p ac ice.
In addi ion o he EQ-5D-3L QoL sco es, which a e
ha d o p edic wi h common eg essions,
82,83
he
DEFENSE su ey
14
assessed he cos s o pa ien s wi h
MS in Finland. The EDSS- ela ed di ec cos s we e
es ima ed based on da a om he DEFENSE su ey
14
and a e epo ed in Table III. Because o limi a ions in
he assessmen o DEFENSE-de i ed elapse cos s, he
cos s o elapses we e es ima ed om o he pa ien s
wi h RRMS in Finland (Tampe e; N ¼581; da a
included p ocedu es, hospi al isi s, hospi al s ays,
and uni cos
70
) using semilog mul i a ia e
me hodology explained elsewhe e.
47,84
Based on his
analysis, he addi ional cos s pe elapse wi h and
wi hou hospi aliza ion we e €5537.57 and €1297.41,
espec i ely.
In a scena io analysis, he ela ionship be ween
EDSS and annual di ec ca e cos s (excluding DMT
cos s) was es ima ed based on a nonlinea in e pola-
ion o findings epo ed in a s udy om Finland,
13
as
ollows:
Annual di ec ðDMTs excl:Þcos s
¼€ð128:44 EDSS2þ4266:60 EDSS–2480:10Þ;
(Equa ion3)
con e ed o 2014 eal alue
71
and wi h EDSS 0 se o
€0. The cos s applied in his sensi i i y analysis we e
well in line wi h hose om o he MS cos s udies
om Finland
15
and elsewhe e.
9–11
Apa om he d ugs, which we e alued a
Janua y 2016 p ices,
66
heal h ca e cos s we e alued
a 2013–2014 eal p ices. The equi ed infla ion
adjus men s we e pe o med using Finland’so ficial
p ice index o communal heal h ca e expendi u es o
income index.
71,85
The modeled cos s and heal h
ou comes we e discoun ed a 3%/y.
Clinical The apeu ics
542 Volume 39 Numbe 3
Sensi i i y and Gene alizabili y o Resul s
The obus ness and gene alizabili y o he base case
esul s we e assessed using a ious de e minis ic and
p obabilis ic sensi i i y analyses (DSA and PSA, e-
spec i ely). The base case was based on mos c edible
inpu s. DSAs we e based on 25 di e en scena ios,
including majo o nonc edible changes in me hods,
heal h isks, ea men , cos s, QoL, popula ion, and
se ings. Means based on all 25 DSA scena ios we e
also calcula ed. The de ails o he DSAs a e shown in
Table IV.
P obabilis ic Sensi i i y Analysis
Fo PSA, a second-o de Mon e Ca lo simula ion
was used o ake in o accoun he join a ia ion in he
economic and clinical ou comes due o sampling
unce ain y ela ed o model pa ame e s. The ollow-
ing dis ibu ions we e used: β o ARR and wi h-
d awal a es, γ o EDSS- ela ed and ea men cos s,
log-no mal o EDSS ansi ions, disease p og ession
haza d a es, ea men e ec on ARR, ea men
e ec on hospi aliza ion elapse pe cen age and
QoL, and Di ichle dis ibu ion o he pe cen age o
elapses in ol ing hospi aliza ion (see Supplemen al
Ma e ial C in he online e sion a h p://dx.doi.o g/
10.1016/j.clin he a.2017.01.028). Based on he PSA,
cos -e ec i eness accep abili y on ie s demons a ed
op imal ea men o maximize ne benefi wi h di e -
en WTP h esholds, and Bayesian ea men anking
anked he bes ea men s.
RESULTS
The a e age modeled base case esul s a e epo ed in
Table V. The mean p ojec ed 15-yea o al paye ’s
di ec cos s di e ed conside ably (by 17.2%) be ween
he mos a o dable ( e iflunomide) and he mos
cos ly (IFN-β1b SC) DMT. The espec i e ela i e
QALY gain di e ence was 9.3%. The maximum
ela i e QALY di e ence was 10.6% be ween he 2
DMTs (DMF and IFN-β1b SC).
The modeled key ou come (ICERs €/QALY gained in
compa ison wi h BSC alone) anged conside ably, om
Table II. D ug- ela ed use and cos s.
DMT Dose/Amoun pe Package
Cos pe
Package,€
*
Dosage (SPCs) Use, % Cos , €
DMF 120 mg
†
120 mg, 14 able s 188.37 120 mg PO BID 1.92 14,435/1s y
DMF 240 mg
†
240 mg, 56 able s 1151.56 240 mg PO BID 15.33
240 mg, 168 able s 3319.33 82.75
DMF 240 mg
†
240 mg, 56 able s 1151.56 240 mg PO BID 15.33 14,523/2nd y
240 mg, 168 able s 3319.33 84.67
GA 20 mg
‡
20 mg/mL, 28 1 mL 836.11 20 mg SC once daily 100.00 10,907
IFN-β1a 30 mgIM
§
30 mg/0.5 mL, 4 0.5 mL 814.90 30 mg SC QW 100.00 10,630
IFN-β1a 44 mgSC
‖
44 mg/0.5 mL, 12 0.5 mL 897.83 44 mg SC TIW 100.00 11,712
IFN-β1b 250 mgSC
¶
250 mg/mL, 15 1 mL 793.08 250 mg SC EOD 100.00 9656
Te iflunomide 14 mg
#
14 mg, 28 able s 1017.89 14 mg PO once daily 15.33 12,023
14 mg, 84 able s 2712.79 84.67
DMT ¼disease-modi ying he apy; DMF ¼dime hyl uma a e; GA ¼gla i ame ace a e; IFN ¼in e e on; SPC ¼summa y
o p oduc cha ac e is ics.
*
D ug cos s a e a Janua y 2016 alues.
†
T adema k: Tecfide a
s
(Biogen, Wes on, Massachuse s).
‡
T adema k: Copaxone
s
(Te a, Ulm, Ge many).
§
T adema k: A onex
s
(Biogen).
‖
T adema k: Rebi
s
(EMD Se ono, Rockland, Massachuse s).
¶
T adema k: Be a e on
s
(Baye Pha maceu icals, Wes Ha en, Connec icu ).
#
T adema k: Aubagio
s
(Genzyme [a SanofiCompany], Camb idge, Massachuse s).
E. Soini e al.
Ma ch 2017 543
24,081 ( e iflunomide) o 248,652 (GA) pe QALY
gained, and BSC domina ed IFN-β1b SC in he base
case. Te iflunomide was es ima ed o be less cos ly and
mo e e ec i e (dominan ) han injec able fi s -line
DMTs, and DMF had a high ICER o 75,431 e sus
e iflunomide, esul ing om he ma ginally mo e QA-
LYs (0.089) wi h DMF and highe cos s e sus e i-
flunomide o e 15 yea s. (Table V and Figu e 3).
I he WTP h eshold o addi ional QALY gained
is se o he mos plausible le el (€25,000), only
e iflunomide ep esen s a cos -e ec i e al e na i e o
BSC alone, based on he modeling. I he WTP is
be ween €37,000 (plausible) and €55,000 (end o li e)
pe QALY gained, only e iflunomide and DMF
ep esen cos -e ec i e al e na i es o BSC alone.
Howe e , wi h a modeled ICER o 75,414 o DMF
e sus e iflunomide, DMF is unlikely o be consid-
e ed cos -e ec i e in he Finnish se ing gi en he
uno ficial assumed WTP h esholds de ailed in Ma e-
ials and Me hods.
The cos –benefi analysis ype IIA u ilized he
minimal mean expec ed DMT- ela ed discoun ed
Table III. Moni o ing and disabili y (EDSS)- ela ed esou ce use and cos s.
Moni o ing Uni Cos , €
*
Resou ces, Fi s Yea /La e Yea
†
DMF GA IFNs Te iflunomide BSC
Specialis isi 340.76, Including 5%
copaymen
69
2/1 2/1 2/1 2/1 0/0
SC aining 50.97 Nu se isi
69
0/0 1/0 1/0 0/0 0/0
Labo a o y ee
‡
5.47
68
4/4 0/0 4/1 17/6 0/0
ALT 1.00
67
4/1 0/0 4/1 17/6 0/0
GGT, c ea inine 2.00
67
4/1 0/0 0/0 0/0 0/0
BC 1.55
67
0/0 0/0 4/1 0/0 0/0
FBC 6.60
67
4/4 0/0 0/0 4/1 0/0
MxA 92.50
70
0/0 0/0 1/1 0/0 0/0
TSH 2.50
67
0/0 0/0 1/0 0/0 0/0
UT 5.84
68
4/1 0/0 0/0 0/0 0/0
MRI, head 335.58
69
1/0.5 1/0.5 1/0.5 1/0.5 0/0
Phone call
§
9.56 A e es s
69
2/3 0/0 2/0 15/5 0/0
Disabili y ela ed
EDSS sco e
14
–0/1 2/3 4/5 6/7 8–9
Di ec heal h
ca e cos s, €
‖
–1108/1446 2890/3470 3909/5656 7919/12,185 15,718
Di ec non–heal h
ca e cos s, €
–49/834 1693/4526 5767/15,289 18,749/32,364 68,852
ALT ¼alanine amino ans e ase; BC ¼blood coun ; BSC ¼bes suppo i e ca e; DMF ¼dime hyl uma a e; EDSS ¼
Expanded Disabili y S a us Scale; FBC ¼ ull blood coun ; GA ¼gla i ame ace a e; GGT ¼gamma-glu amyl ans e ase;
MRI ¼magne ic esonance imaging; MxA ¼p o ein induced by in e e on-al a/β;TSH¼ hy oid-s imula ing ho mone; UT
¼u ine es .
*
P e-2013 non a i cos s
14,68,69
we e indexed o he 2014 p ice le el using o ficial communal p ice index o heal h ca e
se ices.
71
†
Unless o he wise no ed.
‡
Fixed labo a o y ee o each es aking ime.
§
Phone call a e labo a o y es s i specialis isi no a anged.
‖
Es ima ed cos s o disease-modi ying he apies (DMTs) we e excluded based on he digi aliza ion and es ima ion o DMT
cos s in Figu e 4 in Ruu iainen e al.
14
Clinical The apeu ics
544 Volume 39 Numbe 3
budge pe pa ien (minimum WTI) o €42,077 based
on he d ug- ela ed cos s o IFN-β1a SC. The con-
sequen discoun ed IIs in e ms o inc emen al
quali y-adjus ed su i als e sus BSC we e:
e iflunomide, 0.337 QALYs gained; DMF, 0.314;
IFN-β1a SC, 0.264; GA, 0.120; IFN-β1a IM, 0.119;
and IFN-β1b SC, –0.239, all wi h he assumed WTI.
The espec i e inc emen al ime o cane uses we e
Table IV. De ails o de e minis ic sensi i i y analyses.
Ca ego y Scena io
Discoun ing No discoun ing
Discoun ing wi h 5%/y
Heal h isks B i ish Columbia, Canada, RRMS EDSS de elopmen , based on pa ien s mo e han 28 yea s
old
57
Al e na i e na u al elapse sou ce
86
Ra e o elapses leading o hospi aliza ion based on he 1:2.75 a io om Tampe e da a
(26.7% o annual elapses esul in hospi aliza ion when adjus ing o co a ia es including
also EDSS sco e; N ¼581; mean age a elapse, 40 y)
Relapse ime, 2 mo
Relapse ime, 4 mo
T ea men DMT discon inua ion when EDSS 7 and o e was eached, based on eimbu semen c i e ia
Disabili y p og ession and ARR se o he lowe 95% c edibili y in e al h eshold o MTC
esul s
Disabili y p og ession and ARR se o he highe 95% c edibili y in e al h eshold o MTC
esul s
Al e na i e sou ce disabili y p og ession, ARR, and wi hd awal a es om he MTC: no yea
limi and adjus men o placebo elapses
Al e na i e sou ce disabili y p og ession, ARR, and wi hd awal a es om he MTC: no yea
limi
Time wi h AEs doubled (same as doubling AE disu ili y o hose AEs ha las a sho e ime
han he model cycle)
Time wi h AEs hal ed (same as hal ing AE disu ili y)
Cos s EDSS cos s based on he o he Finnish sou ce
13
a 2014 alues
61
Moni o ing cos s doubled
Moni o ing cos s hal ed
Relapse cos doubled
Relapse cos hal ed
AE cos s doubled
AE cos s hal ed
Socie al app oach (p oduc i i y loss included)
14,85
QoL Al e na i e EDSS QoL sou ce
10
Simila QoL loss assumed o all elapses
14
Resul
gene alizabili y
TEMSO
30,32,33
pa ien cha ac e is ics and placebo ansi ion p obabili ies o RRMS EDSS
AE =ad e se e en ; ARR =annualized elapse a e; EDSS =Expanded Disabili y S a us Scale; MTC =mixed- ea men
compa ison; QoL =quali y o li e; RRMS = elapsing– emi ing mul iple scle osis; TEMSO =Randomized T ial o O al
Te iflunomide o Relapsing Mul iple Scle osis) O al Te iflunomide o Pa ien s wi h Relapsing Mul iple.
E. Soini e al.
Ma ch 2017 545
SUPPLEMENTAL MATERIAL
Supplemen al ma e ial accompanying his a icle can
be ound in he online e sion a h p://dx.doi.o g/
10.1016/j.clin he a.2017.01.028.
REFERENCES
1. Finnish Medical Socie y Duodecim, Finnish Neu ological
Socie y Wo king G oup. MS Disease. Cu en Ca e
C i e ia. Helsinki, Finland: Cu en Ca e; 2015.
2. Puglia i M, Rosa i G, Ca on H, e al. The epidemiology
o mul iple scle osis in Eu ope. Eu opean J Neu ology.2006;
13:700–722.
3. Wo ld Heal h O ganiza ion. A las mul iple scle osis
esou ces in he wo ld 2008. Gene a, Swi ze land: Wo ld
Heal h O ganiza ion; 2008:1–56.
4. Holmbe g M, Mu onen A, Elo aa a I, Sumelah i ML.
Inc eased emale MS incidence and di e ences in
gende -specific isk in medium- and high- isk egions in
Finland om 1981-2010. Mul iple Scle osis In e na ional.
2013;2013:182516.
5. Sumelah i ML, Tiena i PJ, Hakama M, Wiks om J.
Mul iple scle osis in Finland: incidence ends and
di e ences in elapsing emi ing and p ima y p og es-
si e disease cou ses. J Neu ol Neu osu g Psychia y. 2003;
74:25–28.
6. Sumelah i ML, Tiena i PJ, Wiks om J, e al. Regional
and empo al a ia ion in he incidence o mul iple
scle osis in Finland 1979-1993. Neu oepidemiology. 2000;
19:67–75.
7. Sumelah i ML, Tiena i PJ, Wiks om J, e al. Inc easing
p e alence o mul iple scle osis in Finland. Ac a Neu ol
Scand. 2001;103:153–158.
8. B une DG, Hopman WM, Singe MA, e al. Measu e-
men o heal h- ela ed quali y o li e in mul iple scle osis
pa ien s. The Canadian J Neu ological Sciences. 1996;23:99–
103.
9. Ka ampampa K, Gus a sson A, an Muns e ET, e al.
T ea men expe ience, bu den, and unme needs (TRIB-
UNE) in Mul iple Scle osis s udy: he cos s and u ili ies
o MS pa ien s in The Ne he lands. J Medical Economics.
2013;16:939–950.
10. Kobel G, Be g J, A he ly D, Hadjimichael O. Cos s and
quali y o li e in mul iple scle osis: a c oss-sec ional
s udy in he Uni ed S a es. Neu ology. 2006;66:
1696–1702.
11. Kobel G, Be g J, Lindg en P, e al. Modeling he cos -
e ec i eness o a new ea men o MS (na alizumab)
compa ed wi h cu en s anda d p ac ice in Sweden.
Mul Scle . 2008;14:679–690.
12. Kobel G, Kas eng F. Access o Inno a i e T ea men s
in Mul iple Scle osis in Eu ope. B ussels, Belgium:
Eu opean Fede a ion o Pha maceu ical Indus ies and
Associa ions; 2009:1–89.
13. Ma ikainen J, Sin onen H. T ea men cos s and quali y o
li e losses. In: Elo aa a I, Pi ilä T, Fä kkilä M, Hie aha ju
A, edi o s. Clinical Neu oimmunology. Helsinki, Finland:
Helsinki Uni e si y P ess Publishing; 2006:184–187.
14.Ruu iainen J, Vii a AM, Hahl J, e al. Bu den o illness in
mul iple scle osis (DEFENSE) s udy: he cos s and
quali y-o -li e o Finnish pa ien s wi h mul iple scle osis.
J Medical Economics. 2016;19:21–33.
15. Sillanpaa M, Andlin-Sobocki P, Lonnq is J. Cos s o
b ain diso de s in Finland. Ac a Neu ol Scand. 2008;117:
167–172.
16. Valjakka S, Nu mi-Koikkalainen P, An ila H, Kon inen
J. Li ing o indi iduals wi h neu ological long- e m illness and
impai men : Helsinki: Aspa Founda ion s ;2013.
17. Sumelah i ML, Hakama M, Elo aa a I, Pukkala E.
Causes o dea h among pa ien s wi h mul iple scle osis.
Mul Scle . 2010;16:1437–1442.
18. Ku zke JF. Ra ing neu ologic impai men in mul iple
scle osis: an expanded disabili y s a us scale (EDSS).
Neu ology. 1983;33:1444–1452.
19. Kingwell E, an de Kop M, Zhao Y, e al. Rela i e
mo ali y and su i al in mul iple scle osis: findings om
B i ish Columbia, Canada. J Neu ol Neu osu g Psychia y.
2012;83:61–66.
20. Poko ski RJ. Long- e m su i al expe ience o pa ien s
wi h mul iple scle osis. J Insu Med. 1997;29:101–106.
21. Biogen Idec. Na alizumab (Tysab i) o he ea men o
adul s wi h highly ac i e elapsing emi ing mul iple
scle osis: Single echnology app aisal (STA) submission
o he Na ional Ins i u e o Heal h and Clinical Ex-
cellence. 2006:1–269.
22. Naci H, Fleu ence R, Bi J, Duhig A. Economic bu den
o mul iple scle osis: a sys ema ic e iew o he li e -
a u e. Pha macoEconomics. 2010;28:363–379.
23. O me M, Ke igan J, Tyas D, e al. The e ec o disease,
unc ional s a us, and elapses on he u ili y o people wi h
mul iple scle osis in he UK. Value Heal h. 2007;10:54–60.
24. P osse LA, Kun z KM, Ba -O A, Weins ein MC. Cos -
e ec i eness o in e e on be a-1a, in e e on be a-1b,
and gla i ame ace a e in newly diagnosed non-p ima y
p og essi e mul iple scle osis. Value Heal h. 2004;7:554–
568.
25. Kelas o. Numbe o indi iduals wi h eimbu semen s
and p esc ip ions. Helsinki: Kela; 2015.
26. Fox RJ, Mille DH, Phillips JT, e al. Placebo-con olled
phase 3 s udy o o al BG-12 o gla i ame in mul iple
scle osis. N Engl J Med. 2012;367:1087–1097.
27. Hu chinson M, Fox RJ, Mille DH, e al. Clinical e ficacy
o BG-12 (dime hyl uma a e) in pa ien s wi h elapsing-
emi ing mul iple scle osis: subg oup analyses o he
CONFIRM s udy. J Neu ol. 2013;260:2286–2296.
Clinical The apeu ics
552 Volume 39 Numbe 3
28. Ba -O A, Gold R, Kappos L,
e al. Clinical e ficacy o BG-12
(dime hyl uma a e) in pa ien s
wi h elapsing- emi ing mul iple
scle osis: subg oup analyses o
he DEFINE s udy. J Neu ol.2013;
260:2297–2305.
29. Gold R, Kappos L, A nold DL,
e al. Placebo-con olled phase 3
s udy o o al BG-12 o elapsing
mul iple scle osis. N Engl J Med.
2012;367:1098–1107.
30. Mille AE, O’Conno P, Wolinsky
JS, e al. P e-specified subg oup
analyses o a placebo-con olled
phase III ial (TEMSO) o o al
e iflunomide in elapsing mul i-
ple scle osis. Mul Scle . 2012;18:
1625–1632.
31. O’Conno P, Comi G, F eedman
MS, e al. Long- e m sa e y and
e ficacy o e iflunomide: Nine-
yea ollow-up o he andomized
TEMSO s udy. Neu ology.2016;
86:920–930.
32. O’Conno P, Wolinsky JS, Con-
a eux C, e al. Randomized ial
o o al e iflunomide o elaps-
ing mul iple scle osis. N Engl
J Med. 2011;365:1293–1303.
33. O’Conno PW, Lublin FD, Wo-
linsky JS, e al. Te iflunomide
educes elapse- ela ed neu olog-
ical sequelae, hospi aliza ions and
s e oid use. JNeu ol. 2013;260:
2472–2480.
34. Con a eux C, O’Conno P,
Comi G, e al. O al e iflunomide
o pa ien s wi h elapsing mul i-
ple scle osis (TOWER): a ando-
mised, double-blind, placebo-
con olled, phase 3 ial. The
Lance . Neu ology. 2014;13:247–
256.
35. Mille AE, Macdonell R, Comi G,
e al. Te iflunomide educes e-
lapses wi h sequelae and elapses
leading o hospi aliza ions: e-
sul s om he TOWER s udy.
J Neu ol. 2014;261:1781–1788.
36. Mille AE, Wolinsky JS, Kappos L,
e al. O al e iflunomide o pa-
ien s wi h a fi s clinical episode
sugges i e o mul iple scle osis
(TOPIC): a andomised, double-
blind, placebo-con olled, phase
3 ial.The Lance . Neu ology.2014;
13:977–986.
37. Ve me sch P, Czlonkowska A, G i-
maldi LM, e al. Te iflunomide
e sus subcu aneous in e e on
be a-1a in pa ien s wi h elapsing
mul iple scle osis: a andomised,
con olled phase 3 ial. Mul Scle .
2014;20:705–716.
38. B iggs A, Clax on C, Sculphe M.
Decision modelling o heal h
economic e alua ion. Ox o d,
UK: Ox o d Uni e si y P ess;
2008.
39. Pha maceu icals P icing Boa d.
Composing heal h economic e al-
ua ion o he p icing and eim-
bu semen applica ion. Helsinki:
Pha maceu icals P icing Boa d.2015.
40. Soini E, Hallinen T, Sokka AL,
Saa inen K. Cos -u ili y o fi s -
line ac inic ke a osis ea men s
in Finland. Ad The . 2015;32:
455–476.
41. Soini E, Kukkonen J, Myllykangas
M, Ryynänen OP. Con ingen
alua ion o eigh new ea -
men s: wha is he clinician’s
and poli ician’s willingness o
pay? The Open Complemen a y
Medicine Jou nal. 2012;4:1–11.
42. Collins M, La ime N. NICE’s end
o li e decision making scheme:
impac on popula ion heal h.
BMJ (Clinical esea ch ed.).2013:
1363.
43. Na ional Ins i u e o Heal h and
Ca e Excellence (NICE). Guide o
he Me hods o Technology Ap-
p aisal 2013. London, UK: NICE;
2013.
44. Hallinen T, Soini E. The impac
o he pha maceu ical p icing
sys em on cos -e ec i eness e-
sul s: Finnish analysis. The Open
Pha macoeconomics & Heal h Eco-
nomics Jou nal. 2011;3:6–10.
45. Hallinen TA, Soini EJ, Eklund K,
Puolakka K. Cos -u ili y o di e -
en ea men s a egies a e he
ailu e o umou nec osis ac o
inhibi o in heuma oid a h i is
in he Finnish se ing. Rheuma ol-
ogy (Ox o d). 2010;49:767–777.
46. Soini E, Hau ala A, Poikonen E,
e al. Cos -e ec i eness o fi s -
line ch onic lymphocy ic leuke-
mia ea men s when ull-dose
fluda abine is unsui able. Clin
The . 2016;38:889–904.
47. Soini E, Leussu M, Hallinen T.
Adminis a ion cos s o in a e-
nous biologic d ugs o heuma-
oid a h i is. Sp inge Plus.2013;
2:531.
48. Soini E, Ma ikainen JA, Nousiai-
nen T. T ea men o ollicula
non-Hodgkin’s lymphoma wi h
o wi hou i uximab: cos -
e ec i eness and alue o in o -
ma ion based on a 5-yea ollow-
up. Annals o oncology: o icial jou nal
o he Eu opean Socie y o Medical
Oncology. 2011;22:1189–1197.
49. Ma ikainen J, Hallinen T, Soini
E. Economic e alua ion o med-
icines. Pha maceu ical econom-
ics and managemen : heo y,
esea ch and p ac ice [in Fin-
nish]. Dosis a maseu inen aika-
kauski ja. 2006;22:289–300.
50. Soini E, Hallinen T, B ignone M,
e al. Cos -u ili y analysis o o -
ioxe ine e sus agomela ine, bu-
p opion SR, se aline and
enla axine XR a e ea men
swi ch in majo dep essi e diso -
de in Finland. Expe Re iew Pha -
Macoeconomics Ou comes Resea ch.
2016:1–10.
51. Peu a P, Ma ikainen J, Soini E,
e al. Cos -e ec i eness o s a ins
in he p e en ion o co ona y
hea disease e en s in middle-
aged Finnish men. Cu Med Res
Opin. 2008;24:1823–1832.
52. Hallinen T, Soini E, Assebu g C,
e al. Cos -e ec i eness o insu-
lin gla gine compa ed o o he
long-ac ing basal insulins in he
ea men o Finnish ype 1 and
ype 2 diabe es pa ien s based
on indi idual s udies. Dosis
E. Soini e al.
Ma ch 2017 553
a maseu inen aikakauski ja.2012;
28:145–164.
53. Soini E. Economic e alua ion o
a e diso de s: wha suppo o
decisions should be p oduced [in
Finnish]? Ha inaise Sai aude .
2011;1:26–31.
54. Soini E. Cos -e ec i eness o a -
h i is ca e in Finland [in Fin-
nish]. Bes P ac ice Reumasai aude
—La
¨a
¨ke ie een Asian un ijoiden
Amma ileh i. 2013;1:7–9.
55. Soini EJ, Da ies G, Ma ikainen
JA, e al. Popula ion-based
heal h-economic e alua ion o
he seconda y p e en ion o co -
ona y hea disease in Finland.
Cu Med Res Opin. 2010;26:25–
36.
56. Fimea. Medicines HTA ad iso y
boa d mee ing memo. Helsinki:
Fimea. 2014:1–3.
57. Palace J, B egenze T, T emle H,
e al. UK mul iple scle osis isk-
sha ing scheme: a new na u al
his o y da ase and an imp o ed
Ma ko model. BMJ open.2014;
4:e004073.
58. Palace J, Duddy M, B egenze T,
e al. E ec i eness and cos -
e ec i eness o in e e on be a
and gla i ame ace a e in he
UK Mul iple Scle osis Risk Sha -
ing Scheme a 6 yea s: a clinical
coho s udy wi h na u al his o y
compa a o . The Lance . Neu ol-
ogy. 2015;14:497–505.
59. Weinshenke BG, Bass B, Rice
GP, e al. The na u al his o y o
mul iple scle osis: a geog aphi-
cally based s udy. I. Clinical
cou se and disabili y. B ain: a
Jou nal o Neu ology. 1989;112:
133–146.
60. Pa zold U, Pockling on PR.
Cou se o mul iple scle osis: fi s
esul s o a p ospec i e s udy
ca ied ou o 102 MS pa ien s
om 1976-1980. Ac a Neu ol
Scand. 1982;65:248–266.
61. S a is ics Finland. Dea hs: S a is-
ics Finland [in Finnish]. Hel-
sinki: S a is ics Finland; 2016.
62. Cu e G, Fah bach K, Huelin R,
e al. Rela i e e ficacy o e iflu-
nomide 14 mg in elapsing MS: a
mixed- ea men compa ison.
O lando: Conso ium o Mul i-
ple Scle osis Cen e s - Ame icas
Commi ee o T ea men and
Resea ch in Mul iple Scle osis;
2013.
63. Na ional Ins i u e o Heal h and
Ca e Excellence (NICE). Final
App aisal De e mina ion—Te iflu-
nomide o T ea ing Relapsing-
Remi ing Mul iple Scle osis.
London, UK: NICE; 2013:1–62.
64. Cohen JA, Coles AJ, A nold DL,
e al. Alem uzumab e sus in e -
e on be a 1a as fi s -line ea -
men o pa ien s wi h elapsing-
emi ing mul iple scle osis: a
andomised con olled phase 3
ial. Lance (London, England).
2012;380:1819–1828.
65. Haas J, Ha ung HP, on Rose-
ns iel P, e al. Fingolimod e-
duces he numbe o se e e
elapses in pa ien s wi h elaps-
ing mul iple scle osis: esul s
om phase III TRANSFORMS
and FREEDOMS s udies. Lisbon,
Po ugal: P esen ed a : Eu opean
Neu ological Socie y; 28-31 May
2011:28–31. Pos e 902.
66. Pha maceu icals P icing Boa d.
Reimbu sable o Au ho ized Me-
dicinal P oduc s [in Finnish].
Helsinki: Pha maceu icals P icing
Boa d; 2016.
67. FimLab. P icelis o Labo a o y
Tes s. Tampe e: FimLab; 2013.
68. Hujanen T, Kapiainen S, Tuomi-
nen U, Peku inen M. Heal h ca e
uni cos s in yea 2006 in Fin-
land. Helsinki: Na ional Ins i u e
o Heal h and Wel a e: STAKES
Wo king Pape s; 2008.
69. Kapiainen S, Väisänen A, Haula
T. Heal h and Social Ca e Cos s
in Yea 2011 in Finland. Helsinki,
Finland: Na ional Ins i u e o
Heal h and Wel a e; 2014.
70. Hospi al Tampe e Uni e si y. Se -
ices and P ices, 2013. Tampe e,
Finland: Tampe e Uni e si y Hos-
pi al; 2013.
71. S a is ics Finland. P ice Index o
Public Expendi u es. Helsinki:
S a is ics Finland; 2016.
72. Soini E, Hallinen T. Cos -u ili y
o agomela ine, enla axine and
placebo in he ea men o ma-
jo dep essi e diso de (MDD)
in Finland—economic modelling
s udy using ep esen a i e popu-
la ion da a. Value Heal h. 2009;
12:A359.
73. Chilco J, McCabe C, Tappen-
den P, e al. Modelling he cos
e ec i eness o in e e on be a
and gla i ame ace a e in he
managemen o mul iple scle o-
sis. Commen a y: e alua ing dis-
ease modi ying ea men s in
mul iple scle osis. BMJ (Clinical
esea ch ed.). 2003:522. discus-
sion 522.
74. Gani R, Gio annoni G, Ba es D,
e al. Cos -e ec i eness analyses
o na alizumab (Tysab i) com-
pa ed wi h o he disease-
modi ying he apies o people
wi h highly ac i e elapsing-
emi ing mul iple scle osis in
he UK. Pha macoEconomics. 2008;
26:617–627.
75.Na ional Ins i u e o Heal h and
Ca e Excellence (NICE). Dime hyl
uma a e o ea ing elapsing-
emi ing mul iple scle osis. NICE
echnology app aisal guidance
320. London, UK: NICE; 2014.
76. Ba end eg JJ. The li e able
me hod o hal cycle co ec ion:
ge ing i igh . Medical Decision
Making: an In e na ional Jou nal o
he Socie y o Medical Decision
Making. 2014;34:283–285.
77. Naima k DM, Kabboul NN,
K ahn MD. The hal -cycle co ec-
ion e isi ed: edemp ion o a
kludge. Medical decision making: an
in e na ional jou nal o he Socie y
o Medical Decision Making.2013;
33:961–970.
78. Fleu ence RL, Hollenbeak CS.
Ra es and p obabili ies in
Clinical The apeu ics
554 Volume 39 Numbe 3
economic modelling: ans o ma-
ion, ansla ion and app op ia e
applica ion. Pha macoEconomics.
2007;25:3–6.
79. Saa ni SI, Ha kanen T, Sin onen
H, e al. The impac o 29
ch onic condi ions on heal h-
ela ed quali y o li e: a gene al
popula ion su ey in Finland us-
ing 15D and EQ-5D. Quali y o
Li e Resea ch: An In e na ional Jou -
nal o Quali y o Li e Aspec s o
T ea men , Ca e and Rehabili a ion.
2006;15:1403–1414.
80. Elo aa a I, Soilu-Hanninen M,
Kuusis o H, e al. Magne ic es-
onance imaging o he b ain in
he moni o ing o immune he -
apy o mul iple scle osis. Duo-
decim. 2015;131:1571–1580.
81. Remes A, Ai as L, A ula S, e al.
Upda e on cu en ca e guide-
line: mul iple scle osis. Duodecim.
2015;131:500–501.
82. Soini E, Koskela T, Ryynänen OP.
In e na ional no malized a io
(INR) moni o ing and pe cen
ime in he apeu ic INR ange
(TTR) ha e impac on pa ien ’s
quali y o li e? Applica ion o
be a eg essions in a p ospec i e
3 mon hs se ing. Value Heal h.
2013;16:A535.
83. Soini E, Mäki inne-Kallio N,
Ma ikainen J, e al. Es ima ion
o quali y-adjus ed su i al wi h
mild obs uc i e sleep apnoea and
six quali y o li e (QoL) assess-
men s: compa ison be ween a-
pezoid ule, c oss-sec ional and
mixed model es ima es. Oslo,
No way: P esen ed a : 14 h Bien-
nial Socie y o Medical Decision
Making (SMDM) Eu opean Mee -
ing. June 10-12, 2012. Abs ac
158.
84. Hallinen T, Ma ikainen JA, Soini
EJ, e al. Di ec cos s o wa a in
ea men among pa ien s wi h
a ial fib illa ion in a Finnish
heal h ca e se ing. Cu
Med Res Opin. 2006;22:683–
692.
85. S a is ics Finland. Income index.
Helsinki: S a is ics Finland; 2016.
86. Held U, Heigenhause L, Shang
C, e al. P edic o s o elapse a e
in MS clinical ials. Neu ology.
2005;65:1769–1773.
87. Bell C, G aham J, Ea nshaw S,
e al. Cos -e ec i eness o ou
immunomodula o y he apies
o elapsing- emi ing mul iple
scle osis: a Ma ko model based
on long- e m clinical da a. Jou -
nal o Managed Ca e Pha macy.
2007;13:245–261.
88. Dembek C, Whi e LA, Quach J,
e al. Cos -e ec i eness o injec -
able disease-modi ying he apies
o he ea men o elapsing
o ms o mul iple scle osis in
Spain. The Eu opean Jou nal o
Heal h Economics: Heal h Economics
in P e en ion and Ca e. 2014;15:
353–362.
89. Imani A, Goles ani M. Cos -u ili y
analysis o disease-modi ying
d ugs in elapsing- emi ing mul i-
ple scle osis in I an. I anian J
Neu ology. 2012;11:87–90.
90. Janko ic SM, Kos ic M, Rado-
sa lje ic M, e al. Cos -
e ec i eness o ou immunomo-
dula o y he apies o elapsing-
emi ing mul iple scle osis: a
Ma ko model based on da a a
Balkan coun y in socioeconomic
ansi ion. Vojnosani e ski P egled.
2009;66:556–562.
91. Pa kin D, Jacoby A, McNamee P,
e al. T ea men o mul iple scle-
osis wi h in e e on be a: an
app aisal o cos -e ec i eness
and quali y o li e. J Neu ol
Neu osu g Psychia y. 2000;68:
144–149.
92. Comi G, F eedman MS, Kappos L,
e al. Pooled sa e y and ole abili y
da a om ou placebo-con olled
e iflunomide s udies and ex en-
sions. Mul iple scle osis and ela ed
diso de s. 2016;5:97–104.
93. CoyleP,Kha iB,Edwa dsK,
e al. Te iflunomide eal-wo ld
sa e y p ofile: esul s o he phase
4 Te i-PRO s udy. P esen ed a :
32nd Cong ess o he Eu opean Com-
mi ee o T ea men and Resea ch in
Mul iple Scle osis. London, Sep em-
be 14-17, 2016. Pos e P648.
94. Gold R, Kha i B, Edwa ds K,
e al. Impac o e iflunomide
ea men on eal-wo ld quali y
o li e in he phase 4 Te i-PRO
s udy. P esen ed a : 32nd Cong ess
o he Eu opean Commi ee o
T ea men and Resea ch in Mul iple
Scle osis. London, Sep embe
14-17, 2016. Pos e P647.
95. Gold R, Kha i B, Edwa ds K,
e al. S able Disabili y and
pa ien - epo ed pe o mance
ou comes o e 48 weeks o e i-
flunomide ea men : esul s
om he phase 4 Te i-PRO s udy.
P esen ed a : 32nd Cong ess o he
Eu opean Commi ee o T ea men
and Resea ch in Mul iple Scle osis.
London, Sep embe 14-17, 2016.
Pos e P646.
96. Moisse X, Zuel M, Conde S,
e al. Te iflunomide and dime-
hyl uma a e: p esc ip ion and
pa ien s’sa is ac ion in a eal li e
se ing. P esen ed a : 32nd Con-
g ess o he Eu opean Commi ee
o T ea men and Resea ch in Mul i-
ple Scle osis. London, Sep embe
14-17, 2016. Pos e P650.
97. Guge M, Enzinge C, Leu meze
F, e al. Real Li e Use o na ali-
zumab, fingolimod, dime hyl u-
ma a e, e iflunomide and
alem uzumab in Aus ia:
benefi - isk da a om he Aus-
ian Mul iple Scle osis T ea -
men Regis y. P esen ed a : 32nd
Cong ess o he Eu opean Commi ee
o T ea men and Resea ch in Mul i-
ple Scle osis. London, Sep embe
14-17, 2016. Pos e EP1510.
98. Anno azzi P, Mallucci G, Lo Re
M, e al. Real li e e ficacy and
ole abili y o e iflunomide: a
mul icen e s udy. P esen ed a :
32nd Cong ess o he Eu opean
Commi ee o T ea men and Re-
sea ch in Mul iple Scle osis.
E. Soini e al.
Ma ch 2017 555
London, Sep embe 14-17, 2016.
Pos e P616.
99. Di Tommaso V, Mancinelli L, Di
Ioia M, e al. E ficacy and sa e y
o fi s line o al he apies in
elapsing- emi ing mul iple scle-
osis: dime hyl uma a e s e i-
flunomide in he Chie i
expe ience. P esen ed a : 32nd
Cong ess o he Eu opean Commi ee
o T ea men and Resea ch in Mul i-
ple Scle osis. London, Sep embe
14-17, 2016. Pos e P1648.
100. Alnajashi H, Alsham ani F, Bakd-
ache F, F eedman M. Di e ences
in ole abili y and discon inua ion
a es in e iflunomide- ea ed pa-
ien s: a eal wo ld clinic expe i-
ence. P esen ed a : 32nd Cong ess o
he Eu opean Commi ee o T ea -
men and Resea ch in Mul iple Scle-
osis. London, Sep embe 14-17,
2016. Pos e P658.
101. Cohan SL, Edwa ds K, Chen C,
e al. Rebound disease ac i i y
educ ion in elapsing mul iple
scle osis pa ien s ansi ioned
om na alizumab o e ifluno-
mide. P esen ed a : 32nd Cong ess
o he Eu opean Commi ee o T ea -
men and Resea ch in Mul iple Scle-
osis. London, Sep embe 14-17,
2016. Pos e P655.
102. Edwa ds K, O’Conno J, Siu a J.
E alua ion o swi ching o e i-
flunomide in high isk na alizu-
mab pa ien s. P esen ed a : 32nd
Cong ess o he Eu opean Commi ee
o T ea men and Resea ch in Mul i-
ple Scle osis. London, Sep embe
14-17, 2016. Pos e P706.
103. Eu opean Medicines Agency. Di-
me hyl uma a e [summa y o
p oduc cha ac e is ics]. 2014.
104. US Food and D ug Adminis a-
ion. Dime hyl uma a e [p e-
sc ibing in o ma ion]. 2016.
105. Soini E, Holmbe g M, Assebe g
C, Sumelah i ML. Modelling he
pe sis ence o disease-modi ying
d ug ea men (DMT) and i s
independen d i e s in Finnish
mul iple scle osis (MS) pa ien s:
pa ame ic su i al modelling.
Value Heal h. 2014;17:A400.
106. Wal he EU, Hohl eld R. Mul iple
scle osis: side e ec s o in e -
e on be a he apy and hei
managemen . Neu ology. 1999;53:
1622–1627.
107. Nik a S, Keb iaeezadeh A, Dina -
and R, e al. Cos -e ec i eness
o di e en in e e on be a p od-
uc s o elapsing- emi ing and
seconda y p og essi e mul iple
scle osis: decision analysis based
on long- e m clinical da a and
swi chable ea men s. Da u: jou -
nal o Facul y o Pha macy, Teh an
Uni e si y o Medical Sciences.2013;
21:50.
108. Rubio-Te és C, A ís egui Ruiz I,
Medina Redondo F, Izquie do
Ayuso G. Cos -u ili y analysis o
mul iple scle osis ea men wi h
gla i ame ace a e o in e e on
be a in Spain. Fa m Hosp. 2003;
27:159–165.
109. Rubio-Te es C, Dominguez-Gil
Hu le A. Cos -u ili y analysis o
elapsing- emi ing mul iple scle-
osis ea men wi h aza hiop ine
o in e e on be a in Spain. Re
Neu ol. 2005;40:705–710.
110. Sanchez-de la Rosa R, Saba e E,
Casado MA, A oyo R. Cos -
e ec i eness analysis o disease
modi ying d ugs (in e e ons and
gla i ame ace a e) as fi s line
ea men s in emi ing- elapsing
mul iple scle osis pa ien s. Jou nal
o Medical Economics. 2012;15:424–
433.
111. Zhang X, Hay JW, Niu X. Cos
e ec i eness o fingolimod, e iflu-
nomide, dime hyl uma a e and
in amuscula in e e on-be a1a in
elapsing- emi ing mul iple scle o-
sis. CNS D ugs.2015;29:71–81.
112. Häkkinen U, I e sen T, Pel ola M,
e al. Heal h ca e pe o mance
compa ison using a disease-based
app oach: he Eu oHOPE p ojec .
Heal h Policy.2013;112:100–109.
113. Linna M, Hakkinen U, Pel ola M,
e al. Measu ing cos e ficiency in
he No dic hospi als—a c oss-
sec ional compa ison o public
hospi als in 2002. Heal h Ca e
Manag Sci. 2010;13:346–357.
114. Medin E, Hakkinen U, Linna M,
e al. In e na ional hospi al p o-
duc i i y compa ison: expe iences
om he No dic coun ies. Heal h
Policy. 2013;112:80–87.
115. Nyblin K. In which di ec ion is
he eimbu semen sys em o
pha maceu icals going o e ol e?
De enso legis. 2009;6:915.
116. Minis y o Social A ai s and
Heal h, Minis y o Finance.
Heal hca e, social wel a e and
egional go e nmen e o m pack-
age. Helsinki, Finland: Minis y o
Social A ai s and Heal h, Minis-
y o Finance; 2016.
117. Soini E. Expe solu ions in ou -
comes esea ch. Legal Wo king
G oup Wo kshop, Seconda y
Use o Da a: In o ma ion Man-
agemen , Con ol and Moni o ing
[in Finnish]. Helsinki, Finland:
Minis y o Social A ai s and
Heal h; 2016.
118. Pelkonen L. Condi ional Reim-
bu semen [in Finnish]. Helsinki:
Finnish Pha maceu icals P icing Boa d.
2016.
119. Riksdag Pa liamen . Go e nmen
p oposal 184/2016 p [in Swed-
ish]. 2016.
120. Belgian Heal h Ca e Knowledge
Cen e. Belgian Guidelines o
Economic E alua ions and Budge
Impac Analyses. B ussels: Belgian
Heal h Ca e Knowledge Cen e;
2012.
121. College o Heal h Economis s.
F ench guidelines o he eco-
nomic e alua ion o heal h ca e
echnologies [in F ench]. Pa is:
College o Heal h Economis s;
2004:1–90.
122. Heal h Ca e Insu ance Boa d.
Guidelines o pha macoeco-
nomic esea ch, upda ed e sion
[in Du ch]. Diemen, The Ne he -
lands: Heal h Ca e Insu ance
Boa d; 2006:1–14.
Clinical The apeu ics
556 Volume 39 Numbe 3
123. Heal h In o ma ion and Quali y
Au ho i y. Guidelines o he eco-
nomic e alua ion o heal h ech-
nologies in I eland. Co k, I eland:
Heal h In o ma ion and Quali y
Au ho i y; 2010:1–75.
124. Ins i u e o Pha maceu ical Re-
sea ch. Guidelines on Heal h Eco-
nomic E alua ion [in Ge man].
Vienna, Aus ia: Ins i u e o Pha -
maceu ical Resea ch; 2006:1–11.
125. Bal ic guideline o economic
e alua ion o pha maceu icals
(pha macoeconomic analysis).
La ia, Es onia and Li huania: 8
Augus 2002. In e na ional Soci-
e y o Pha macoeconomics and
Ou comes Resea ch; 2002:1–6.
126. López-Bas ida J, Oli a J,
An onanzas F, e al. Spanish
ecommenda ions on economic
e alua ion o heal h echnologies.
The Eu opean Jou nal o Heal h
Economics: Heal h Economics in P e-
en ion and Ca e. 2010;11:513–
520.
127. No wegian Medicines Agency.
Guidelines on How To Conduc
Pha macoeconomic Analyses [in
No wegian]. Oslo: No wegian
Medicines Agency; 2012:1–27.
128. Sco ish Medicines Conso ium.
Guidance o Manu ac u e s o
Comple ion o New P oduc As-
sessmen Fo m. Glasgow: Sco ish
Medicines Conso ium; 2016.
129. Soini E, Hallinen T, Kauppi M,
e al. Comp ehensi e heal h eco-
nomic assessmen o sequenced
ea men wi h biologics in
mode a e- o-se e e heuma oid
a h i is: analysis based on
ACR50 and ACR70 esponses.
A h i is Rheum. 2010;62:759.
130. Soini E, Ma ikainen JA, Vihe -
aa a V, e al. Economic e alua-
ion o sequen ial ea men s o
ollicula non-Hodgkin lym-
phoma. Clin The . 2012;34:e912.
131. Shi ani A, Zhao Y, Kingwell E,
e al. Tempo al ends o disabil-
i y p og ession in mul iple
scle osis: findings om B i ish
Columbia, Canada (1975-2009).
Mul Scle . 2012;18:442–450.
132. T emle H, Zhao Y, De onshi e
V. Na u al his o y compa isons
o p ima y and seconda y p o-
g essi e mul iple scle osis e eals
di e ences and simila i ies. J Neu-
ol. 2009;256:374–381.
133. Con a eux C, Vukusic S, Adeleine
P. Ea ly clinical p edic o s and
p og ession o i e e sible disabili y
in mul iple scle osis: an amnesic
p ocess. B ain: A Jou nal o Neu ol-
ogy. 2003;126:770–782.
134. Le ay E, Yaouanq J, Le Page E,
e al. E idence o a wo-s age
disabili y p og ession in mul iple
scle osis. B ain: a Jou nal o Neu-
ology. 2010;133:1900–1913.
135. Minden SL, F ankel D, Hadden L,
e al. The Sonya Sli ka longi udi-
nal mul iple scle osis s udy:
me hods and sample cha ac e -
is ics. Mul Scle . 2006;12:24–38.
Add ess co espondence o: M . E kki Soini, ESiOR L d, Tullipo inka u
2 LT4, Kuopio 70100, Finland. E-mail: e kki.soini@esio .fi
E. Soini e al.
Ma ch 2017 557
SUPPLEMENTARY MATERIAL
Supplemen A. EDSS-based RRMS and SPMS
ansi ion ma ices
EDSS
1
is he key ou come in he assessmen o MS
disabili y p og ession. In he Finnish Pi kanmaa-
Seinäjoki-Vaasa MS egis y, he e we e 1359 pa ien s
wi h MS wi h EDSS assessmen da a a ailable, wi h
al oge he 2458 measu emen s. These pa ien s we e
iden ified om adminis a i e egis ies. The da a
collec ion, case asce ainmen p ocedu e, and e hical
pe mi s ha e been desc ibed in de ail elsewhe e.
2,3
Inciden MS cases diagnosed in he s udy egion ha
ulfilled he McDonald
4
c i e ia we e included. The
classifica ion o disease cou se o RRMS was
pe o med using s anda dized defini ions.
5
A o al o 1242 pa ien s had RRMS, and hese
pa ien s wi h RRMS had al oge he 2299 EDSS
measu emen s be ween Augus 27, 1986, and Decem-
be 31, 2010. Women accoun ed o 69.8% o he
pa ien s. In all, 62.2% o he EDSS assessmen s we e
ca ied ou a he beginning o a DMT episode wi h an
EDSS sco e o 0–7. In 2010, EDSS alues we e
assessed o all pa ien s ali e (July 1, 2010, assumed,
i no specific day shown in he da a).
EDSS T ansi ions in RRMS
Figu e A.1 shows all EDSS measu emen s o e ime
o desc ip i e pu poses. As can be seen, mos EDSS
measu emen s we e pe o med o pa ien s wi h
RRMS (g een colo ed do s). The figu e also shows
ha he e was censo ing in he EDSS measu emen s in
EDSS classes 6.5–9.5.
Fo desc ip i e pu poses, combined Figu e A.2
shows he de elopmen om one EDSS measu emen
o he nex among pa ien s wi h RRMS, condi ional
on pa icula EDSS sco es.
EDSS de elopmen o e ime needs o be modeled
in o de o es ima e he p og ession o MS. MS
p og ession o he model was es ima ed using in ege
RRMS EDSS sco es (hal es ounded up; 9.5 assumed
o be 9.0 because he pa ien is ali e when EDSS is
9.5). The JAGS so wa e V3.3.0,
6
which is a s a is ical
p og am capable o analyzing Bayesian hie a chical
models by Ma ko Chain Mon e Ca lo (MCMC)
0
1
1.5
2
2.5
3
3.5
4
4.5
5
5.5
6
6.5
7
7.5
8
8.5
9
10
1990 1995 2000 2005 2010
Calenda ime
EDSS (ji e ed)
Sub ype
PPMS
RRMS
SPMS
All EDSS measu emen s o e ime
Figu e A.1. All EDSS measu emen s (ji e ed o p e en o e -plo ing) o e ime, by MS ype (RRMS, g een;
PPMS, ed; SPMS, blue). EDSS ¼Expanded Disabili y S a us Scale; PPMS ¼p ima y p og essi e
mul iple scle osis; RRMS ¼ elapsing- emi ing mul iple scle osis; SPMS ¼seconda y p og essi e
mul iple scle osis.
Clinical The apeu ics
557.e1 Volume 39 Numbe 3
RRMS: nex EDSS, cu en EDSS is 0
RRMS: nex EDSS, cu en EDSS is 1.5 o 2
RRMS: nex EDSS, cu en EDSS is 3.5 o 4
RRMS: nex EDSS, cu en EDSS is abo e 5.0
RRMS: nex EDSS, cu en EDSS is 4.5 o 5
RRMS: nex EDSS, cu en EDSS is 2.5 o 3
RRMS: nex EDSS, cu en EDSS is 1
10.0
7.5
5.0
2.5
0.0
10.0
7.5
5.0
2.5
0.0
10.0
7.5
5.0
2.5
10.0
7.5
5.0
2.5
0.0
05Y s 10 0510 15
Y s
10.0
7.5
5.0
2.5
0.0
0510 15
10.0
7.5
5.0
2.5
0.0
0 5 Y s 10 15
Y s Y s
Y s
Y s
0510 15
0510
15 0510 15 20
10.0
7.5
5.0
2.5
0.0
EDSS(ji e ed) EDSS(ji e ed)
EDSS(ji e ed) EDSS(ji e ed)
EDSS(ji e ed)EDSS(ji e ed)
EDSS(ji e ed)
Figu e A.2. EDSS de elopmen in RRMS popula ion o e ime showing nex EDSS sco es. The blue line gi es
he a e age expec ed EDSS o e ime and he shaded a ea is he 95% CI ob ained by unadjus ed
local polynomial smoo hing o he aw da a. CI ¼con idence in e al; EDSS ¼Expanded
Disabili y S a us Scale; RRMS ¼ elapsing- emi ing mul iple scle osis.
E. Soini e al.
Ma ch 2017 557.e2
Table A.I. Annual ansi ion p obabili y ma ix by EDSS o pa ien s wi h RRMS based on he Finnish da a.
F om/To
RRMS
EDSS 0
RRMS
EDSS 1
RRMS
EDSS 2
RRMS
EDSS 3
RRMS
EDSS 4
RRMS
EDSS 5
RRMS
EDSS 6
RRMS
EDSS 7
RRMS
EDSS 8
RRMS
EDSS 9
RRMS
EDSS 10
RRMS
EDSS 0
0.67822 0.26314 0.04275 0.01136 0.00364 0.00077 0.00003 0.00003 0.00003 0.00003 0.00000
RRMS
EDSS 1
0.11299 0.60711 0.17922 0.06484 0.02725 0.00711 0.00037 0.00037 0.00037 0.00037 0.00000
RRMS
EDSS 2
0.01770 0.17312 0.37521 0.22712 0.14263 0.04960 0.00365 0.00365 0.00365 0.00365 0.00001
RRMS
EDSS 3
0.00547 0.07282 0.26542 0.25690 0.24007 0.11065 0.01216 0.01216 0.01216 0.01216 0.00002
RRMS
EDSS 4
0.00155 0.02710 0.14772 0.21289 0.30097 0.18210 0.03189 0.03189 0.03189 0.03189 0.00009
RRMS
EDSS 5
0.00045 0.00981 0.07124 0.13607 0.25204 0.20969 0.08010 0.08010 0.08010 0.08010 0.00031
RRMS
EDSS 6
0.00001 0.00027 0.00276 0.00786 0.02314 0.04173 0.23071 0.23071 0.23071 0.23071 0.00141
RRMS
EDSS 7
0.00001 0.00027 0.00276 0.00786 0.02314 0.04173 0.23071 0.23071 0.23071 0.23071 0.00141
RRMS
EDSS 8
0.00001 0.00027 0.00276 0.00786 0.02314 0.04173 0.23071 0.23071 0.23071 0.23071 0.00141
RRMS
EDSS 9
0.00001 0.00027 0.00276 0.00786 0.02314 0.04173 0.23071 0.23071 0.23071 0.23071 0.00141
RRMS
EDSS 10
0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 1.00000
Clinical The apeu ics
557.e3 Volume 39 Numbe 3
simula ion me hods, was used o es ima e he EDSS
ansi ion p obabili ies.
When es ima ing he RRMS EDSS 0–9 ansi ions,
uni o m p io s we e assumed because no ea lie
Finnish ansi ion p obabili ies da a we e a ailable.
Based on a p io knowledge o he da a in ques ion,
60% o he mo ali y was assumed o be MS- ela ed.
7
This es ima e was conse a i e in compa ison o o he
es ima es, which ha e a highe p opo ion o MS-
ela ed mo ali y (eg, 78.3% in Goodin e al
8
). The
esul s shown in Table A.I a e well in line wi h he
ecen B i ish Columbia esul s.
9
RRMS o SPMS T ansi ion
The haza d a e (HR, λ) o con e sion om
RRMS EDSS 1 o SPMS was calcula ed assuming an
exponen ial su i al unc ion (ie, a cons an haza d o
con e ing o SPMS o e ime):
Sð Þ¼expðλ Þ
λ o an exponen ial dis ibu ion could be es ima ed
om he median ime o con e sion o SPMS, e-
po ed o be 15 yea s based on London On a io
da a,
10,13
ie:
λ¼lnð2Þ=15
This gi es an annual HR o 0.0462 o SPMS-
con e sion o pa ien s in EDSS 1.
The Finnish da ase includes only a ew obse a-
ions o con e sion o SPMS, and an EDSS-specific
a e could no be es ima ed om hese. Based on he
London On a io da a, he Cox p opo ional haza ds
model was:
Hð Þ¼Hð ÞEDSS1:expðβXÞ
whe e H( ) is he HR o con e sion o any EDSS
s a e; H( )
EDDS1
, he HR o con e sion o EDDS 1;
and β, he coe ficien (0.25270) o he ela ionship
be ween EDSS and he HR o p og ession be ween he
base case EDSS 1 and all o he EDSS s a es.
10,13
Using
Bende e al,
11
he ela ionship was e o mula ed as:
ln Hð Þ
Hð ÞEDSS1
¼β:X
Thus:
Hð Þ¼λ:eβ:X
This was used o de i e he HR o con e sion om
EDSS 1 h ough each successi e s age o EDSS 8
(Table A.II). All es ima ed HRs we e hen
subsequen ly con e ed in o p obabili ies
12
:
p¼1expð Þ
EDSS T ansi ions in SPMS
Fo SPMS ansi ions, da a om he London
On a io MS egis y
10,13
we e a ailable and used
(Table A.III), because he Finnish egis e da a had
oo ew EDSS measu emen s o pa ien s wi h SPMS.
Table A.II. Annual p obabili ies o con e sion o SPMS om RRMS by EDSS sco e.
EDSS sco e Calcula ion Haza d a e o con e sion Calcula ion P obabili y
1 ln(2)/15 0.046210 1-exp(-0.046210) 0.045158
2 0.04621*e (0.25270*2) 0.076600 1-exp(-0.076600) 0.073739
3 0.04621*e (0.25270*3) 0.098622 1-exp(-0.098622) 0.093915
4 0.04621*e (0.25270*4) 0.126975 1-exp(-0.126975) 0.119245
5 0.04621*e (0.25270*5) 0.163480 1-exp(-0.163480) 0.150817
6 0.04621*e (0.25270*6) 0.210480 1-exp(-0.210480) 0.189805
7 0.04621*e (0.25270*7) 0.270993 1-exp(-0.270993) 0.237378
8 0.04621*e (0.25270*8) 0.348902 1-exp(-0.348902) 0.294538
9
†
1.000000
10 0.000000
†
In o ma ion o EDSS 9 was no a ailable om he London On a io da ase . Thus, a 100% con e sion a e o pa ien s wi h
RRMS in EDSS 9 was assumed.
E. Soini e al.
Ma ch 2017 557.e4