1275
In oduc ion
Venous h omboses cause high mo ali y in cance pa ien s
[1, 2]. P io s udies conduc ed mainly on pedia ic acu e
lymphoblas ic leukemia (ALL) pa ien s epo associa ions
o enous h omboses (VTs) wi h pa ien - (high body mass
index (BMI), ad anced age, como bidi ies), disease- (ALL
sub ype), and ea men - ela ed ac o s (aspa aginase, s e -
oids, and in a hecal chemo he apy) [3–10]. Mos VTs
including ce eb al enous h omboses (CVTs) in ALL
pa ien s occu du ing he i s 2 o 3 mon hs o leukemia
ea men [3–5].
CVTs co e up o o e 30% o all cases o VTs in
ALL pa ien s. I leads o d ama ic consequences such as
ORIGINAL RESEARCH
A mino ole o aspa aginase in p edisposing o ce eb al
enous h omboses in adul acu e lymphoblas ic leukemia
pa ien s
Saa a Roininen1,2 , Ou i Laine3,4, Ma ju Kauppila5, Ma ko Vesanen5, Ma ia Räme 3, Ma ja a
Sinisalo3, Esa Jan unen6, Ma jaana Säily7, Riikka Rä y1,2, E kki Elonen1 & Ulla Wa io aa a-Kau o1,2
1Comp ehensi e Cance Cen e , Depa men o Hema ology, Helsinki Uni e si y Hospi al, Helsinki, Finland
2Uni e si y o Helsinki, Helsinki, Finland
3Depa men o In e nal Medicine, Tampe e Uni e si y Hospi al, Tampe e, Finland
4Uni e si y o Tampe e, Tampe e, Finland
5Depa men o In e nal Medicine, Tu ku Uni e si y Hospi al, Tu ku, Finland
6Depa men o In e nal Medicine, Kuopio Uni e si y Hospi al, Kuopio, Finland
7Depa men o In e nal Medicine, Oulu Uni e si y Hospi al, Oulu, Finland
© 2017 The Au ho s. Cance Medicine published by John Wiley & Sons L d.
This is an open access a icle unde he e ms o he C ea i e Commons A ibu ion 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.
Keywo ds
Acu e lymphoblas ic leukemia, aspa aginase,
ce eb al enous h ombosis, isk ac o s
Co espondence
Ulla Wa io aa a-Kau o, Helsinki Uni e si y
Hospi al, Comp ehensi e Cance Cen e ,
Depa men o Hema ology, Uni e si y o
Helsinki, Finland. PO box 372, 00029 HUS,
Helsinki, Finland. Tel: +35894711;
Fax: +358947177351;
E-mail: [email p o ec ed]
Funding In o ma ion
Resea ch g an s om Weikko Wilhelm
Pel onen Founda ion (2014), and he Finnish
Haema ological Resea ch Founda ion (2016)
and he Signe and Ane Gyllenbe g
Founda ion (2016), and a el g an s om
he Eu opean Haema ology Associa ion
(2016) and he Uni e si y o Helsinki Fund
(2016).
Recei ed: 9 Decembe 2016; Re ised: 14
Ma ch 2017; Accep ed: 11 Ap il 2017
Cance Medicine 2017; 6(6):1275–1285
doi: 10.1002/cam4.1094
Abs ac
Ce eb al enous h ombosis (CVT) co e s up o a hi d o all enous h om-
boses (VTs) de ec ed in pa ien s wi h acu e lymphoblas ic leukemia (ALL). I
usually hampe s pa ien s’ li es and may also endange e icien leukemia ea -
men . Al hough many ac o s ha e been sugges ed o accoun o an ele a ed
isk o VTs in pa ien s wi h ALL, he e s ill is a lack o s udies ocusing on
CVTs and especially in he se ing o adul ALL pa ien s. We s udied in ou
e ospec i e popula ion- based coho he occu ence, cha ac e is ics, as well as
isk ac o s o VTs in 186 consecu i ely diagnosed Finnish adul ALL pa ien s
ea ed wi h a na ional pedia ic- inspi ed ea men p o ocol ALL2000. In he
isk ac o analyses o VTs we ound a dis inc ion o he cha ac e is ics o he
pa ien s acqui ing CVT om hose wi h o he kinds o VTs o wi hou h om-
bosis. In con as o p e ious s udies we we e also able o compa e he e ec s
o aspa aginase in ela ion o CVT occu ence. No ably, mo e han hal o he
CVTs we e diagnosed p io he adminis a ion o aspa aginase which accen ua es
he ole o o he isk ac o s on he pa hophysiology o CVT compa ed o
uncal o cen al enous line (CVL) VTs in adul ALL pa ien s.
Cance
Medicine Open Access
1276 © 2017 The Au ho s. Cance Medicine published by John Wiley & Sons L d.
S. Roininen e al.Mino Role o Aspa aginase in CVT in Adul ALL
epilepsy and cogni i e o ocal de ici s in a signi ican
amoun o encoun e ed pa ien s [3, 6]. E iological ac o s
iden i ied in nonleukemia pa ien coho s include emale
sex, ho monal manipula ion (e.g., o al con acep i es),
ce ain malignancies, and head auma, bu hey explain
less han hal o he CVT cases [11, 12]. Al hough CVT
is a ela i ely common complica ion in ALL pa ien s, we
s ill do no know he speci ic biological basis o his e en
[3, 10].
CVTs may be ha d o diagnose because he symp oms
a y om mild headaches o li e- h ea ening seizu es o
nausea wi h concomi an in ac anial hype ension. The
usual ime lag be ween he onse o symp oms and diag-
nosis is abou 2 weeks [11, 12].
Aspa aginase is conside ed e icien and essen ial in ALL
ea men . I deple es ee aspa agine and glu amine om
he ex acellula luid. This leads o apop osis o he malig-
nan lymphoblas ic cells as hei cell g ow h and di ision
a e highly dependen o hese ci cula ing amino acids [13].
Aside om he aimed an ileukemic p ope ies, aspa aginase
exposes pa ien s o mul iple ad e se e ec s such as VTs.
O he suspec ed, al hough no unanimously iden i ied, isk
ac o s o VTs in ALL pa ien s include high BMI, T- cell
ALL (T- ALL), and he concomi an use o s e oids [14–17].
As o ou knowledge, epo s on whe he he isk ac o s
o CVTs and uncal o cen al enous line (VCL) VTs
in ALL pa ien s a e alike, s ill lack.
We pe o med a popula ion- based e ospec i e egis y
s udy on he Finnish ALL pa ien s ea ed wi h a na ional
ALL2000 s udy p o ocol du ing 2000–2012. We aimed o
disco e he po en ial dis inc ion o he isk ac o s o
CVTs and o he VTs in ALL pa ien s. Kho ana sco e has
been used o de ini ion o pa ien s in high isk o enous
h ombosis in solid umo malignancies [18]. In his s udy
we also aimed o e ospec i ely moni o , whe he basic
labo a o y es s and mo phological ac o s de ec able a
ALL diagnosis could e eal pa ien s in a high isk o a
CVT.
Pa ien s and me hods
The s udy ini ially comp ised o 201 consecu i e adul
ALL pa ien s (aged 16–65 yea s) ea ed wi h a na ional
p o ocol ALL2000 ( esul s unpublished; MEA s. CVAD
as induc ion, CVAD as he i s consolida ion, and aspa a-
ginase in oduced in he second consolida ion; Table 1)
in he i e Finnish uni e si y hospi als (Helsinki, Kuopio,
Oulu, Tampe e, and Tu ku) and Vaasa Cen al Hospi al
du ing 1999–2012. Pa ien s we e ollowed up o Ap il
2015 (median o al ollow- up: 4.2 yea s, ange: 0.02–
14.9 yea s) a e which he analysis was conduc ed. Pa ien
consen was ob ained acco ding o he Decla a ion o
Helsinki and he s udy was app o ed by he espec i e
E hics commi ee o each pa icipa ing hospi als. O 201
pa ien s, 186 we e included in ou analysis and 15 excluded
by ollowing c i e ia: insu icien pa ien eco ds (n = 11),
alse p ima y diagnose (one lymphoma and one ch onic
lymphoblas ic leukemia), and p io acu e leukemia (n = 2)
(Fig. S1).
In he ALL2000 s udy pa ien s we e andomized in o
wo di e en induc ion g oups, CVAD and MEA, a e
which hey ecei ed iden ical consolida ion he apies
(Table 1). CVAD was gi en o 96 (52%) and MEA o
90 (48%) o he pa ien s. Esche ichia coli- de i ed aspa a-
ginase was in oduced a 8 h day o he second consoli-
da ion block. In ou s udy he median in e al om he
ALL diagnosis o he i s ac ualized dose o aspa aginase
was 64 days ( ange: 43–196 days). Follow- up o he
occu ence o VTs (median: 88 days, ange: 7–219 days)
was he ime measu ed om leukemia diagnosis o he
end o second consolida ion block, allogeneic s em cell
ansplan a ion (alloSCT), o dea h. P ima y endpoin s o
ou s udy we e a VT, alloSCT, o dea h. Only wo o
ou s udy pa ien s ecei ed p ophylac ic an icoagula ion
du ing leukemia ea men . Thei indica ion o p ophylaxis
was a ial ib illa ion.
The clinical pheno ypes and basic labo a o y alues o
he pa ien s we e eco ded and included in ou analysis.
De ailed desc ip ion o hese pa ame e s is shown in
Table 2 and Table S1. All da a we e collec ed om pa ien
cha s. Only VTs con i med by adiological imaging ech-
niques (ul asound, compu ed omog aphy, o magne ic
esonance imaging) we e included in he analysis. Pa ien
cha ac e is ic a he ime o ALL diagnosis (Table 2),
clinical cha ac e is ics o he VT pa ien s (Table 3), and
he da a used in he analysis (Table S1) a e shown.
S a is ical analyses
SAS s a is ical so wa e 9.3 (SAS Ins i u e, Ca e, NC) and
IBM SPSS s a is ics 22 we e used o he s a is ical analysis.
Mann–Whi ney U o Pea son Chi- squa ed (χ2) es s we e
used in uni a ia e analysis o pa ien cha ac e is ics
depending on he da a cha ac e is ics. Cox eg ession model
was plo ed wi h ime- dependen co a ia e (aspa aginase
ea men ) and he isk sco e model o CVT by using
FREQ p ocedu e. All analyses we e wo sided and P < 0.05
was conside ed s a is ically signi ican .
Resul s
In o al, 31 (17%) pa ien s su e ed om a VT, o whom
nine (27%) had a CVT (Table 3). O he ypes o VTs
(n = 22) included en lowe and one uppe ex emi y
deep enous h omboses, six cen al enous- line h om-
boses, and i e pulmona y embolisms. The exac da es o
1277
© 2017 The Au ho s. Cance Medicine published by John Wiley & Sons L d.
Mino Role o Aspa aginase in CVT in Adul ALLS. Roininen e al.
all VTs a e shown on Table 1. CVTs occu ed mos ly a
he i s consolida ion he apy which was ea lie han o he
eco ded VTs (median: 43 days, ange: 34–126 and median:
68 days, ange: 1–127 and, espec i ely; P > 0.05).
A de ailed desc ip ion o he pa ien cha ac e is ics and
dis ibu ion o hem among leukemia o h ombosis sub-
ypes is gi en in Table 3. Dis ibu ion o CVAD o MEA
induc ion among pa ien s su e ing om h ombosis du ing
he ollow- up is shown in Table 1.
The uni a ia e analysis showed a signi ican ly lowe
median BMI (21.5 kg/m2, ange: 17.0–26.0 kg/m2) o CVT
pa ien s compa ed o pa ien s wi h o he VTs (P = 0.002)
o wi hou VTs (P = 0.018), espec i ely. Fu he mo e,
CVT pa ien s had mo e T- ALL (5/9, 56%) and ex amed-
ulla y leukemia (5/9, 56%) han pa ien s wi h o he VTs
(3/22, (14%), P = 0.007 and 4/22, (18%), P = 0.037)
and no VTs (23/155, (15%), P = 0.005 and 30/155, (19%),
P = 0.010) (Table 3). CVT pa ien s had a signi ican ly
highe CRP compa ed o pa ien s wi h o he VTs
(P = 0.033), bu a di e ence in leukocy e and pla ele
coun s did no each s a is ical signi icance (Table 3).
De ailed cha ac e is ics o pa ien s wi h CVTs a e shown
in Table 4. Mos o he CVTs (5/9 (56%)) occu ed be o e
he i s adminis a ion o aspa aginase and we did no
de ec associa ion o aspa aginase wi h CVT in he Cox
eg ession model (χ2: 0.000, P = 0.995). Lowe BMI and
T- ALL sub ype emained signi ican ac o s in also his
analysis (χ2: 3.926, P = 0.048 and χ2: 4.8708, P = 0.025,
espec i ely). O he CVT pa ien s, 8/9 (89%) had been
on o al s e oids (dexame hasone), bu none had ecei ed
in a hecal chemo he apy (cy a abine o me ho exa e) a
week p io he h ombosis.
Pa ien s wi h uncal o CVL VTs exp essed ypical
gene al isk cha ac e is ics o VTs in he uni a ia e
analysis. They we e olde , had mo e como bidi ies, and
had a signi ican ly highe median BMI (27.1 kg/m2,
ange: 20.2–39.5, P = 0.038) compa ed o pa ien s wi h-
ou VTs. (Table 3). Only 5/22 (32%) o he pa ien s
su e ed om h ombosis be o e he i s dosing o
aspa aginase and he Cox eg ession analysis showed a
clea associa ion o aspa aginase ea men wi h occu -
ence o uncal o CVL VTs (χ2: 6.850, P = 0.0089).
Fu he mo e, 21/22 (95%) ecei ed o al s e oids and
3/22 (14%) in a hecal ea men less han a week p io
o he VT.
Ex amedulla y leukemia was ound in 44 o 186 (24%)
pa ien s, and dis ibu ion o ex amedulla y leukemia is
shown in Table S2. O he nine CVT pa ien s, wo had
a medias inal umo mass, wo ma kedly enla ged lymph
nodes, and one leukemic skin in il a ion a diagnosis.
Nine y- eigh (53%) o all pa ien s, 15 (48%) o pa ien s
wi h o he VTs, and 5 (56%) o pa ien s wi h a CVT
we e ali e a he ime he da a we e analyzed (Table S3).
Mos o all VTs (in o al 52%) we e diagnosed and ea ed
in Helsinki Uni e si y Hospi al (Table S4). No s a is ically
signi ican di e ences be ween he incidence a es o VTs
we e de ec ed be ween he hospi als.
We also cons uc ed a p elimina y CVT isk sco e model
based on ou esul s, p io knowledge o CVT isk ac-
o s, and Kho ana isk sco e model [11, 12, 18]. Age,
gende , BMI, disease sub ype, leukemia dissemina ion,
hemoglobin, and pla ele coun s a ALL diagnosis we e
included in he sco e. Ranking limi s (0 o 1 poin /pa am-
e e ) we e selec ed acco ding o he associa ion s udy
esul s p e iously published [18]. A isk sco e o ≥5 poin s
equals o a high isk o CVT (Fig. 1). Pa ien s in he
high- isk g oup ep esen ed mo e han 20- old isk o
CVT compa ed o he pa ien s in he low- isk g oup
(haza d a io = 20.841, P < 0.0001, 95% CI: 5.208–83.401).
The sco e model showed an excellen speci ici y and a
mode a e sensi i i y (NPV: 0.982, PPV: 0.375).
Discussion
Venous h omboses o en comp omise he e ec i e ea -
men o ALL. We pe o med a de ailed analysis o he
occu ence and isk ac o s o VTs in a consecu i e se ies
o adul pa ien s diagnosed in Finland in 2000–2012 and
ea ed acco ding o a na ional s udy p o ocol ALL2000.
The incidence a es o VTs we e compa able o p e ious
epo s [10, 12, 15–17]. Risk ac o s o uncal o CVL
VTs we e as hypo hesized and e lec ed many o he a-
di ional haza ds o h ombosis de ec ed in he gene al
popula ion. These ac o s, howe e , we e dis inc om
hose de ec ed in pa ien s wi h CVT in ou s udy coho .
Namely, pa ien s acqui ing CVT we e younge and had
a s a is ically signi ican ly lowe BMI and highe incidences
o ex amedulla y and T- cell leukemia compa ed o pa ien s
wi h no VTs and uncal o CVL VTs. A s udy made by
Zuu bie e al. showed no di e ences in age (median 33
s. 33 yea s) be ween CVT and no CVT pa ien s, whe eas
CVT pa ien s in a s udy conduc ed by Cou u ie e al.
showed age dis ibu ion (median 29 s. 33 yea s) nea
o ou esul s bu wi hou a s a is ical signi icance [14,
15]. In compa ison o ou esul s hese disc epancies may
be explained by a di e en age dis ibu ion o he pa ien
popula ion (16–65 and 16–18 o 59 yea s, espec i ely).
Young age as isk ac o o CVT is suppo ed by he
e iew by McBane e al., whe e median o he pa ien s
ecei ing CVT we e a he age o ea ly o ies [11]. The
dis inc i ely low BMI in CVT pa ien s in ou analysis is
likely explained by he young age o he a ec ed pa ien s.
CVT is a se e e complica ion ha may lead o di icul
sequelae [11, 12, 14]. I o en causes delays o unca ions
in he ea men o ALL. Hence, we hink ha clinicians
should be awa e ha no only he o e weighed and olde
1278 © 2017 The Au ho s. Cance Medicine published by John Wiley & Sons L d.
S. Roininen e al.Mino Role o Aspa aginase in CVT in Adul ALL
pa ien s a e a isk o a se ious h ombo ic complica ions.
As delays in diagnoses o CVTs a e equen ou esul s
may also ca y on impac in sensi izing clinicians o o de
p ope adiological in es iga ions o pa ien s wi h neu-
ological symp oms.
Medias inal mass may inc ease he isk o h ombosis
by, o example, inc easing medias inal p essu e ha could
impai enous low om he ce eb al eins [6]. Two (22%)
o he nine CVT pa ien s (Table 4) had a medias inal
mass a ALL diagnosis in ou s udy. Due o he limi ed
numbe o he s udy pa ien s, he e iological ole o hese
masses in CVT occu ence canno be es ima ed. Compa ed
o he ALL- sub ype dis ibu ion o all ALL2000 pa ien s
(Table 2 and Table 3) ou CVT pa ien s had 3.3- old
incidence o T- ALL. This is mo e han epo ed p e iously
(2- and 1.5- old) [14, 15]. T- ALL is mo e equen in
males which likely explains he e en dis ibu ion o ou
CVT cases be ween gende s in ou analysis. T- ALL is also
Table 1. Occu ence o enous h omboses in he ALL2000 cy os a ic ea men . This able shows he cy os a ic p o ocol and he ime o enous
h ombosis (VT) diagnosis in he i s h ee ea men blocks o he ALL2000 s udy. In a hecal cy os a ic and aspa aginase ea men s a e
highligh ed wi h colo s (ligh o ange and ligh ed, espec i ely). Pa ien s wi h enous h omboses (VTs, numbe s ep esen pa ien codes) a e also
highligh ed wi h colo s ha a e explained in he oo no e.
Cy os a ic agen s 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 Cumula i e doses/m²
MEA induc ion
Mi oxan one i. . ↑↑↑↑ 60 mg
E oposide i. . ↑↑↑↑ 400 mg
Cy a abine i. . ↑↑ ↑↑ ↑↑ ↑↑ 8000 mg/4000 mg
Me ho exa e i. . 1± ↑1± ↑425 mg
Cy a abine i. . ↑475 mg
CVAD induc ion
Cyclophosphamide i. . ↑↑ ↑↑ ↑↑ ↑↑ ↑ 3600 mg
Vinc is ine i. . ↑ ↑ 42 mg
Doxo ubicin i. . ↑50 mg
DEXA p.o. ↑ ↑ ↑ ↑ ↑ ↑ ↑ ↑ 160 mg
Me ho exa e i. . 1± ↑1± ↑425 mg
Cy a abine i. . ↑475 mg
VTs 4081,3
4553
4561,2
Fi s cons. he apy
Cyclophosphamide i. . ↑↑ ↑↑ ↑↑ ↑↑ ↑ 3600 mg
Vinc is ine i. . ↑ ↑ 42 mg
Doxo ubicin i. . ↑50 mg
DEXA p.o. ↑ ↑ ↑ ↑ ↑ ↑ ↑ ↑ 160 mg
Me ho exa e i. . ↑412.5 mg
Cy a abine i. . ↑475 mg
VTs 521,2 3511,3 11031191,3 115312131221,2
1651,2
24 28
Second cons. he apy
Dauno ubicin i. . ↑↑↑ 180 mg/135 mg
Vinc is ine i. . ↑ ↑ ____ ____ ____ ____ 2.8 mg/m² + 1.6 mg
DEXA p.o. ↑ ↑ ↑ ↑ ↑ ↑ ↑ ↑ 160 mg
Aspa aginase i. . ↑ ↑ ↑ ↑ ↑430 000 IU/20 000 IU
Cy a abine i. . ↑475 mg
VTs 1321,2 52625121,3 30721311,3 135221322312
1531,3 1621,3 60125821302511244931091,2
4201,3 4581,2 1571,3 3142
In case o wo di e en doses, he i s one o pa ien s <56 yea s and he las one o pa ien s >56 yea s o age. 1 ± : 1 dose gi en in one o
hese 3 days; ↑: one dose daily; ↑↑: wo doses daily. , Ce eb al enous h ombosis; , Lowe ex emi y h ombosis; , Cen al enous- line
h ombosis; , Uppe ex emi y h ombosis; , Pulmona y embolism.
Cons., consolida ion.
1Pa ien s who we e dead a he ime he da a we e analyzed.
2CVAD.
3MEA.
4D ug gi en only, when pla ele coun > 40 × 10ˆ9.
1279
© 2017 The Au ho s. Cance Medicine published by John Wiley & Sons L d.
Mino Role o Aspa aginase in CVT in Adul ALLS. Roininen e al.
ecu en ly associa ed wi h an in e io ou come o ALL
[19]. These ac o s—age, BMI, ex amedulla y leukemia,
and T- ALL— oge he wi h he p echemo he apy end o
low pla ele and high leukocy e coun s migh e lec an
agg essi e o m o leukemia ha po en ially and speci i-
cally p edisposes pa ien s o CVTs. Un o una ely, nei he
de ailed genomic da a on inhe i ed ac o s a ec ing pa ien s’
po en ial h ombophilia we e a ailable no biobank samples
collec ed due o echnical limi a ions in he molecula
gene ic diagnos ics du ing he e a o ALL2000 da a
collec ion.
Con a y o cu en consen , o e hal (56%) o he
pa ien s su e ed om CVT be o e he in oduc ion o
aspa aginase in ou s udy coho [10]. In addi ion, wo
o he nine CVT pa ien s su e ed om p od omal symp-
oms be o e he i s dose o aspa aginase (p oblems wi h
gai n = 1, pe sis en headaches n = 1) (Table 3). This
could ha e e lec ed an al eady ongoing h ombo ic
Table 1. Occu ence o enous h omboses in he ALL2000 cy os a ic ea men . This able shows he cy os a ic p o ocol and he ime o enous
h ombosis (VT) diagnosis in he i s h ee ea men blocks o he ALL2000 s udy. In a hecal cy os a ic and aspa aginase ea men s a e
highligh ed wi h colo s (ligh o ange and ligh ed, espec i ely). Pa ien s wi h enous h omboses (VTs, numbe s ep esen pa ien codes) a e also
highligh ed wi h colo s ha a e explained in he oo no e.
Cy os a ic agen s 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 Cumula i e doses/m²
MEA induc ion
Mi oxan one i. . ↑↑↑↑ 60 mg
E oposide i. . ↑↑↑↑ 400 mg
Cy a abine i. . ↑↑ ↑↑ ↑↑ ↑↑ 8000 mg/4000 mg
Me ho exa e i. . 1± ↑1± ↑425 mg
Cy a abine i. . ↑475 mg
CVAD induc ion
Cyclophosphamide i. . ↑↑ ↑↑ ↑↑ ↑↑ ↑ 3600 mg
Vinc is ine i. . ↑ ↑ 42 mg
Doxo ubicin i. . ↑50 mg
DEXA p.o. ↑ ↑ ↑ ↑ ↑ ↑ ↑ ↑ 160 mg
Me ho exa e i. . 1± ↑1± ↑425 mg
Cy a abine i. . ↑475 mg
VTs 4081,3
4553
4561,2
Fi s cons. he apy
Cyclophosphamide i. . ↑↑ ↑↑ ↑↑ ↑↑ ↑ 3600 mg
Vinc is ine i. . ↑ ↑ 42 mg
Doxo ubicin i. . ↑50 mg
DEXA p.o. ↑ ↑ ↑ ↑ ↑ ↑ ↑ ↑ 160 mg
Me ho exa e i. . ↑412.5 mg
Cy a abine i. . ↑475 mg
VTs 521,2 3511,3 11031191,3 115312131221,2
1651,2
24 28
Second cons. he apy
Dauno ubicin i. . ↑↑↑ 180 mg/135 mg
Vinc is ine i. . ↑ ↑ ____ ____ ____ ____ 2.8 mg/m² + 1.6 mg
DEXA p.o. ↑ ↑ ↑ ↑ ↑ ↑ ↑ ↑ 160 mg
Aspa aginase i. . ↑ ↑ ↑ ↑ ↑430 000 IU/20 000 IU
Cy a abine i. . ↑475 mg
VTs 1321,2 52625121,3 30721311,3 135221322312
1531,3 1621,3 60125821302511244931091,2
4201,3 4581,2 1571,3 3142
In case o wo di e en doses, he i s one o pa ien s <56 yea s and he las one o pa ien s >56 yea s o age. 1 ± : 1 dose gi en in one o
hese 3 days; ↑: one dose daily; ↑↑: wo doses daily. , Ce eb al enous h ombosis; , Lowe ex emi y h ombosis; , Cen al enous- line
h ombosis; , Uppe ex emi y h ombosis; , Pulmona y embolism.
Cons., consolida ion.
1Pa ien s who we e dead a he ime he da a we e analyzed.
2CVAD.
3MEA.
4D ug gi en only, when pla ele coun > 40 × 10ˆ9.

1280 © 2017 The Au ho s. Cance Medicine published by John Wiley & Sons L d.
S. Roininen e al.Mino Role o Aspa aginase in CVT in Adul ALL
p ocess. Unlike ALL2000 p o ocol, mos cu en ALL ea -
men p o ocols in oduce aspa aginase a induc ion. These
ypes o s udy se ings migh he e o e pa ly o e weigh
he causal ole o aspa aginase o CVTs. Ou esul s
demons a e o he i s ime a mino associa i e ole
o aspa aginase ea men on he occu ence o CVT. The
analysis is unique because i enabled us o s udy he isk
ac o s o an ea ly occu ing CVT in adul ALL pa ien s
wi hou he e ec o aspa aginase du ing he i s wo
ea men blocks whe e he isk o CVT is a he highes .
We sugges ha his no el inding u he s eng hens he
idea o he di e ences in pa hophysiology o CVT, and
uncal o CVL- ela ed h ombosis. We also sugges ha
ALL disease i sel migh play a bigge ole in he de el-
opmen o CVT han hough be o e.
Cen al ne ous sys em (CNS) malignancies and in a h-
ecal chemo he apy may con ibu e o he isk o CVT
de elopmen , by, o example, inc easing he b ain issue
p o h ombin le els, o by damaging he blood–b ain ba -
ie [11, 14]. None o ou pa ien s had ecei ed in a hecal
chemo he apy sho ly be o e he h ombosis (Table 3).
Al hough none o ou CVT pa ien s had a de ec able
CNS leukemia in he mo phological analysis o he ce -
eb ospinal luid, i is s ill possible ha a minimal, occul
CNS leukemia could accoun o an inc eased isk o
CVT. Reg e ably, an i h ombin le els ha ha e been
shown o inc ease he isk o h ombosis we e no ou-
inely collec ed a he ime o pa ien en olmen in his
s udy [9]. A ecen s udy on acu e myeloid leukemia
pa ien s sugges s ha a si ua ion pa allel wi h dissemina ed
in a ascula coagulopa hy (i.e., low pla ele coun , low
TT%, and high D- dime ) could inc ease he isk o enous
h ombosis [20]. Un o una ely, al hough ou CVT
pa ien s showed a dec eased le el o pla ele s and ele a ed
Table 2. Pa ien cha ac e is ics a ALL diagnosis. Pa ien s we e di ided acco ding o ALL disease sub ypes.
Pa ien s All (n = 186) p e- B (n = 118) Ph+ (n = 37) T (n = 31)
Va iables Yes, n (%)/Median
( ange)
Yes, n (%)/Median
( ange)
Yes, n (%)/Median
( ange)
Yes, n (%)/
Median ( ange)
Gende (Male) 111 (60) 71 (60) 20 (54) 21 (65)
Age 40.9 (16.1- 65.8) 39.9 (16.1- 65.8) 48.9 (19.2- 64.6) 35.6 (17.0- 61.5)
BMI 25.5 (17.0- 46.5) 25.5 (17.5- 46.5) 25.7 (19.7- 40.3) 25.0 (17.0- 39.5)
Ex amedulla y leukemia 44 (24) 21 (18) 3 (9) 19 (61)
Abdomen, medias inum, skin, and/
o lymph nodes
39 (21) 18 (16) 2 (5) 19 (61)
CNS leukemia 4 (3) 3 (2) 1 (3) 0 (0)
P io como bidi ies
No como bidi ies 118 (64) 70 (60) 24 (65) 24 (77)
A leas one como bidi y 66 (36) 46 (40) 13 (35) 7 (23)
Hype ension 18 (10) 15 (13) 2 (5) 1 (3)
Hype choles e olemia 12 (6) 10 (8) 2 (5) 0 (0)
Diabe es 10 (5) 7 (6) 2 (5) 1 (3)
As hma 7 (4) 6 (5) 0 (0) 1 (3)
Thy oid diso de 7 (4) 4 (3) 3 (8) 0 (0)
A ial ib illa ion 2 (1) 0 (0) 2 (5) 0 (0)
Myoca dial in a c ion 3 (2) 3 (3) 0 (0) 0 (0)
Epilepsy 3 (2) 2 (2) 0 (0) 1 (3)
Ulce a i e coli is 4 (2) 4 (3) 0 (0) 0 (0)
P e ious cance 7 (4) 4 (3) 3 (8) 0 (0)
P io enous h ombosis 2 (1) 2 (2) 0 (0) 0 (0)
O he ac o s
Smoking (cu en /p io ) 32 (17) 18 (15) 6 (16) 8 (26)
In ec ion less han a week p io o
ALL diagnosis
21 (12) 12 (10) 4 (12) 5 (16)
BMI > 30 31 (17) 23 (21) 4 (11) 4 (13)
Ho monal he apy (p oges in/
es ogen) a diagnosis
6 (3) 3 (3) 1 (3) 2 (6)
An icoagula ion a diagnosis 3 (2) 2 (2) 1 (3) 0 (0)
Philadelphia- posi i e (Ph+) pa ien s we e s a is ically signi ican ly olde han p ecu so B- ALL (p e- B) and T- ALL (T) pa ien s (P = 0.032 and P = 0.009,
espec i ely). T- ALL pa ien s had signi ican ly mo e ex amedulla y leukemia compa ed o o he disease sub ypes (χ2: 31.984, P = 0.001). No u he
s a is ically signi ican di e ences be ween ALL sub ypes we e de ec ed in o he pa ien cha ac e is ics.
BMI, body mass index; CNS, Cen al ne ous sys em.
1281
© 2017 The Au ho s. Cance Medicine published by John Wiley & Sons L d.
Mino Role o Aspa aginase in CVT in Adul ALLS. Roininen e al.
Table 3. Clinical cha ac e is ics o pa ien s wi h enous h omboses.
Type o e en a iables No enous
h ombosis
(n = 155)
All enous
h omboses
(n = 31)
O he enous
h omboses
(n = 22)
CVTs (n = 9) All s. no
enous
h omboses
O he s. no
enous
h omboses
CVTs s. no
enous
h omboses
CVTs s.
o he enous
h omboses
P echemo he apy labo a o y alues Median ( ange) Median ( ange) Median ( ange) Median ( ange) P- alue P- alue P- alue P- alue
Hemoglobin 94.5 (35–172) 106 (39–159) 107 (68–159) 105 (39–136) 0.163 0.265 0.307 0.695
Pla ele s 56.0 (4–445) 56.0 (3–291) 57.5 (7–291) 35.0 (3–204) 0.869 0.697 0.309 0.306
Leukocy es 11.0 (0.6–307) 17.9 (1.3–237) 16.3 (1.3–237) 18.9 (2–221) 0.333 0.642 0.269 0.500
CRP 18.5 (1–419)) 17.0 (2–221) 12.0 (2–139) 43.0 (2–221) 0.863 0.236 0.123 0.033
D- Dime 1.80 (0.02–97) 2.80 (0.1–71) 2.40 (0.2–12) 5.40 (0.1–71) 0.539 0.899 0.318 0.236
B- blas s 4.70 (0–260) 11.6 (0–219) 8.30 (0–219) 13.7 (0–217) 0.431 0.746 0.300 0.497
Pa ien and disease cha ac e is ics
Age 41.2 (16.1–56.4) 40.6 (18.2–65.3) 43.6 (18.2–65.3) 27.1 (18.6–58.2) 0.852 0.275 0.166 0.056
BMI 25.5 (17.5–46.5) 25.3 (17.0–39.5) 27.0 (20.2–39.5) 21.5 (17.0–26.0) 0.673 0.038 0.018 0.002
Gende , Male, n (%) 95 (61) 17 (55) 13 (62) 4 (44) 0.547 0.907 0.335 0.457
Disease subg oups Yes, n (%) Yes, n (%) Yes, n (%) Yes, n (%)
p ecu so B- cell ALL 98 (63) 20 (65) 18 (82) 2 (22)
Philadelphia- posi i e B- cell ALL 34 (22) 3 (14) 1 (5) 2 (22)
T- cell ALL 23 (15) 8 (38) 3 (14) 5 (56) 0.148 0.148 0.005 0.007
Ex amedulla y leukemia 30 (19) 9 (30) 4 (18) 5 (56) 0.227 0.942 0.010 0.037
CNS leukemia 4 (3) 0 (0) 0 (0) 0 (0) 0.366 0.444 0.626
≥ 1 Como bidi ies 51 (33) 15 (48) 12 (55) 3 (33) 0.111 0.053 1.000 0.283
Smoking his o y 24 (16) 8 (26) 6 (27) 3 (33) 0.080 0.183 0.171 0.210
In ec ion a diagnosis 16 (11) 5 (17) 3 (14) 2 (29) 0.327 0.695 0.152 0.362
T ea men - associa ed ac o s a enous h ombosis
Day o he h ombosis 68 (0–217) 43 (34–126) 0.349
Cen al enous- line ca he e a h ombosis 8 (44) 4 (44) 1.000
Aspa aginase be o e h ombosis 15 (68) 4 (44) 0.218
In a hecal chemo he apy a week be o e h ombosis 3 (14) 0 (0) 0.244
Dexame hasone a week be o e h ombosis 21 (95) 9 (100) 0.516
BMI, Body mass index; CVT, Cen al enous h ombosis.
1282 © 2017 The Au ho s. Cance Medicine published by John Wiley & Sons L d.
S. Roininen e al.Mino Role o Aspa aginase in CVT in Adul ALL
Table 4. Cha ac e is ic o pa ien s wi h ce eb al enous h omboses.
Pa ien cha ac e is ics and labo a o y alues a diagnosis Cha ac e is ics o CVTs T ea men - ela ed ac o s
Nb Sex Day^ Disease
sub ype
EM
dis-
ease
CNS
leu-
kemia
Age BMI Pl s Hb Leuk CRP D- dime Symp oms Si e Imaging ASP DEXA^^ I .
he apy^^
1 F 34 Ph+ No No 41.0 23.6 3 98 9.8 186 71 Righ hemipa esis,
e igo, speech de ici 1
SSS, le
ans e se
sinus
MRI Ne e Yes No
2 F 39 p e- B No No 20.8 21.5 27 39 18.9 4 0.1 Headache1SSS CT +
MRI
Ne e Yes No
3 M 43 T Yes No 20.7 19.6 204 125 2 10 0.2 Le hemipa esis,
seizu es, headache1
SSS MRI Ne e Yes No
4 F 43 p e- B Yes No 26.3 20.4 72 105 221 73 5.9 Numbness o he le
a m, ace, and ongue1
Co ical eins MRI ′Ne e Yes No
5 M 73 T Yes No 37.0 23.7 22 85 35.7 23 35 Le hemipa esis and
numbness o he le
side1
SSS MRI Yes Yes No
6 M 50 T Yes No 27.1 21.0 78 120 3.8 131 5.9 Headache1SSS, enous
in a c ions
MRI Yes Yes No
7 F 65 T No No 18.5 17.0 74 113 188 33 5.4 Numbness and pa esis
o he le a m2
SSS CT +
MRI
Yes Yes No
8 F 42 Ph+ No No 48.2 26.0 35 136 64.4 221 0.5 Righ hemipa esis,
speech de ici 2
SSS, co ical
eins,
hemo hage
CT +
MRI
Ne e No No
9 M 126 T Yes No 40.2 25.4 30 104 15 43 3.6 Headache, a igue SSS, sinus
ec us
MRI Yes Yes No
Nb, pa ien numbe ; Day^, numbe o days be ween ALL diagnosis and he CVT; EM, ex amedulla y; CNS, cen al ne ous sys em; BMI, body mass index; Pl , pla ele coun ; Hb, hemoglobin; Leuk,
leukocy es; ASP, aspa aginase; DEXA, dexame hasone; I. ., in a hecal; ^^, less han a week be o e he CVT. Pa ame e s used in he able: F, emale; M, male; Ph+, Philadelphia- posi i e ALL; p e- B,
p ecu so B- ALL; T, T- ALL.
1P od omal symp oms
2P od omal synd omes ha occu ed be o e he in oduc ion o aspa aginase; SSS: supe io sagi al sinus; MRI: magne ic esonance imaging; and CT: compu e omog aphy.
1283
© 2017 The Au ho s. Cance Medicine published by John Wiley & Sons L d.
Mino Role o Aspa aginase in CVT in Adul ALLS. Roininen e al.
le el o D- dime , no TT% was measu ed a he ime o
ALL diagnosis. This idea could no he e o e be e alua ed
in ou coho .
Based on Kho ana isk sco e model, esul s ob ained in
ou s udy, and he knowledge o common CVT isk ac-
o s, we buil a p elimina y isk sco e model o CVT [11,
12, 18]. Re ospec i e s udy design and limi ed numbe o
pa ien s, howe e , es ic ou analysis and i he e o e needs
o be alida ed in a la ge and p ospec i e s udy coho .
Ou s udy is p obably he i s one o analyze di -
e ences in po en ial isk ac o s o a CVT compa ed
wi h pa ien g oups o bo h no VTs and wi h o he
VTs in adul ALL. Su p isingly, aspa aginase ea men
did no show associa ion wi h he occu ence o CVTs
in ou s udy se ing. Pa ien s wi h a CVT we e also
leane and younge han o he subjec s. We sugges ha
a he han impu ed acqui ed isk ac o s o VTs, such
as aspa aginase, he cha ac e is ics o he ALL disease
i sel migh play a signi ican ole in he de elopmen
o CVT in adul ALL pa ien s. We hink ha mo e
s udies aiming a alida ion o an indi idualized CVT
isk es ima ion in ALL pa ien s a e needed o . Gene ic
ac o s ela ed bo h o leukemia and inhe i ed h om-
bophilia should also be in es iga ed and po en ially aken
in o accoun .
Acknowledgmen s
The au ho s would like o hank p o esso s Kimmo Po kka
and Tom Pe e sson om he Uni e si y o Helsinki o
hei in ellec ual inpu , s a is icians Teppo Hu unen and
Joni Ke o- Tokoi om 4Pha ma o hei echnical inpu ,
he Weikko Wilhelm Pel onen Founda ion (2014), he
Finnish Haema ological Resea ch Founda ion (2016), and
he Signe and Ane Gyllenbe g Founda ion (2016) o he
esea ch g an s, and also he Eu opean Haema ology
Associa ion (2016) and he Uni e si y o Helsinki Fund
(2016) o he a el g an s.
Con lic o In e es
None decla ed.
Figu e 1. Risk sco e model o ce eb al enous h ombosis. (A) Shows g aphics o he isk sco e model o ce eb al enous h ombosis (CTV) whe e
CVT pa ien s (ligh g ay ba s, n = 9) a e compa ed o pa ien s wi hou CVT (da k g ay ba s, n = 177) a he ime o ALL diagnosis. Pe cen ages o
amoun o pa ien s belonging o each isk g oup (x- axis) a e shown on he y- axis. O CVT and no CVT pa ien s, 6/9 (67%) and 10/177 (6%) sco ed ≥
5 poin s (high- isk g oup), espec i ely. (B) Shows a iables and cu - o s used in he isk sco e model o CVT. Cu - o s wi h P- alue < 0.05 we e used.
Hemoglobin cu - o was based on Kho ana isk sco e model [15] and emale sex based on he emale dominance in he CVT cases [9].
Va iables Cu -o s Risk alues - alues P- alues
Hemoglobin> 100 g/L1p1.421 0.233
Pla ele coun <80 x 1091p 3.140 0.076
SexFemale1p0.912 0.340
Age<41 yea s1p5.019 0.025
BMI< 4 kg/m21p 5.634 0.018
Disease sub ype T-ALL1p11.141 0.004
Leukemia
dissemina ion
Ex amedulla y
leukemia
1p 5.599 0.018
Risk g oups Abb e ia ion Cu -o
Low- isk LR <5p
High- iskHR5p
A
B
2