Ci a ion: Da Fonseca, L.G.;
Hashizume, P.H.; de Oli ei a, I.S.;
Izquie do-Sanchez, L.; Saud, L.R.d.C.;
Xe an, M.P.; Al es, V.A.F.; de Mello,
E.S.; He man, P.; Banales, J.M.; e al.
Associa ion be ween Me abolic
Diso de s and Cholangioca cinoma:
Impac o a Pos ula ed Risk Fac o
wi h Rising Incidence. Cance s 2022,
14, 3483. h ps://doi.o g/10.3390/
cance s14143483
Academic Edi o s: S en Loosen and
Ch is oph Rode bu g
Recei ed: 25 May 2022
Accep ed: 12 July 2022
Published: 18 July 2022
Publishe ’s No e: MDPI s ays neu al
wi h ega d o ju isdic ional claims in
published maps and ins i u ional a il-
ia ions.
Copy igh : © 2022 by he au ho s.
Licensee MDPI, Basel, Swi ze land.
This a icle is an open access a icle
dis ibu ed unde he e ms and
condi ions o he C ea i e Commons
A ibu ion (CC BY) license (h ps://
c ea i ecommons.o g/licenses/by/
4.0/).
cance s
A icle
Associa ion be ween Me abolic Diso de s and
Cholangioca cinoma: Impac o a Pos ula ed Risk Fac o wi h
Rising Incidence
Leona do G. Da Fonseca 1,2,*,† , Ped o H. Hashizume 1,†, I ai San ana de Oli ei a 3, Lau a Izquie do-Sanchez 4,
Lisa Rod igues da Cunha Saud 2,5, Ma iana Pinhei o Xe an 5, Venancio A ancini Fe ei a Al es 2,6,
E and o Sob oza de Mello 2,6, Paulo He man 2,7, Jesus M. Banales 4,8, Claudia P. Oli ei a 2,5 and Flai J. Ca ilho 2,5
1Clinical Oncology, Ins i u o do Cance do Es ado de São Paulo, School o Medicine, Uni e si y o São Paulo,
São Paulo 01246-000, B azil; [email p o ec ed]
2São Paulo Clínicas Li e Cance G oup—Ins i u o do Cance do Es ado de São Paulo, 255 ICHC—9 h Floo ,
Room 9159, São Paulo 05403-000, B azil; [email p o ec ed] (L.R.d.C.S.);
[email p o ec ed] (V.A.F.A.); [email p o ec ed] (E.S.d.M.); [email p o ec ed] (P.H.);
[email p o ec ed] (C.P.O.); [email p o ec ed] (F.J.C.)
3Depa men o Radiology, Ins i u o do Cance do Es ado de São Paulo, School o Medicine,
Uni e si y o São Paulo, São Paulo 01246-000, B azil; [email p o ec ed]
4Depa men o Li e and Gas oin es inal Diseases, Biodonos ia Heal h Resea ch Ins i u e—Donos ia
Uni e si y Hospi al, Uni e si y o he Basque Coun y (UPV/EHU), CIBERehd, Ike basque,
48009 San Sebas ian
, Spain; [email p o ec ed]g (L.I.-S.); [email p o ec ed]g (J.M.B.)
5Depa men o Gas oen e ology, Di ision o Clinical Gas oen e ology and Hepa ology, School o Medicine,
Hospi al das Clinicas, Uni e si y o São Paulo, São Paulo 05403-000, B azil; [email p o ec ed]
6Depa men o Pa hology, Uni e si y o São Paulo School o Medicine, São Paulo 05403-000, B azil
7Depa men o Gas oen e ology, Di ision o Diges i e Su ge y, School o Medicine, Hospi al das Clinicas,
Uni e si y o São Paulo, São Paulo 04021-001, B azil
8
Depa men o Biochemis y and Gene ics, School o Sciences, Uni e si y o Na a a, 31080 Pamplona, Spain
*Co espondence: [email p o ec ed]
† These au ho s con ibu ed equally o his wo k.
Simple Summa y:
A po en ial ela ionship be ween cholangioca cinoma and me abolic diso de s has
been sugges ed, bu he e is a lack o published da a. This s udy aimed o desc ibe he p e alence o
me abolic diso de s in a coho o 122 pa ien s wi h cholangioca cinoma and epo clinical ou comes.
We ound a p e alence o 42.6% o me abolic diso de s. The e was no signi ican di e ence in o e all
su i al be ween pa ien s wi h o wi hou me abolic diso de s, al hough he e was a be e su i al
in he subg oup o pa ien s unde going su gical esec ion. This indica es a need o be e explo e he
associa ion be ween cholangioca cinoma in a me abolic backg ound.
Abs ac :
In oduc ion and objec i es: The incidence o cholangioca cinoma (CCA) has been in-
c easing globally. Al hough a concomi an inc ease in he incidence o me abolic diso de s migh
sugges a causal ela ionship, he da a a e sca ce. We aimed o desc ibe he p e alence o me abolic
diso de s in pa ien s wi h CCA and epo he clinical ea u es and ou comes. Pa ien s and Me hods:
Re ospec i e s udy including pa ien s wi h CCA. Pa ien s we e di ided in o: (1) pas his o y o dia-
be es o /and o e weigh /obesi y (“me abolic diso de g oup”) and (2) wi hou any o hese ea u es
(“non-me abolic-diso de g oup”). A Cox eg ession model was used o de e mine he p ognos ic
ac o s. Resul s: 122 pa ien s we e included. In o al, 36 (29.5%) had o e weigh /obesi y, 24 (19.7%)
had diabe es, and 8 (6.6%) had bo h. A o al o 29 (23.8%) pa ien s had esec able disease and ecei ed
up on su ge y. A o al o 104 (85.2%) ecei ed chemo he apy o ad anced/ ecu en disease. The
o e all su i al o he coho was 14.3 mon hs (95% CI: 10.1–17.3). ECOG-PS 0 (
p< 0.0001
), esec able
disease (p= 0.018) and absence o ascula in asion (p= 0.048) we e independen ly associa ed wi h
be e p ognosis. The “me abolic diso de g oup” (n= 52) had a median su i al o 15.5 mon hs
(95% CI 10.9–33.9) s. 11.5 mon hs (95% CI 8.4–16.5) in he “non-me abolic-diso de g oup” (n= 70)
(HR: 1.10; 95% CI 0.62–1.94). Pa ien s wi h esec able disease in he “me abolic g oup” had longe
su i al han pa ien s in he “non-me abolic g oup” (43.4 mon hs (95% CI 33.9-NR) s. 21.8 mon hs
Cance s 2022,14, 3483. h ps://doi.o g/10.3390/cance s14143483 h ps://www.mdpi.com/jou nal/cance s
Cance s 2022,14, 3483 2 o 14
(95% CI 8.6–26.9); HR = 0.12, 95% CI 0.03–0.59). Conclusion: Me abolic diso de s a e equen among
CCA pa ien s. Unde lying me abolic como bidi ies may be associa ed wi h p ognosis in esec able
CCA. The e is a need o explo e he mechanism ha d i es CCA ca cinogenesis in a me abolic
backg ound.
Keywo ds: li e cance ; cholangioca cinoma; me abolic synd ome; diabe es; obesi y
1. In oduc ion
Cholangioca cinoma (CAA) ep esen s a he e ogeneous en i y no only because o he
ana omical si e o o igin (in a-hepa ic, pe ihila and dis al CCA) bu also due o di e ences
in molecula ea u es ha a e being inc easingly explo ed in ecen s udies [1].
CCA has been associa ed wi h se e al hepa obilia y diseases, such as p ima y scle-
osing cholangi is (PSC), p ima y bilia y cholangi is (PBC), choleli hiasis, i al hepa i is
in ec ion, li e luke in es a ion, ci hosis and in lamma o y bowel disease. Howe e ,
hese isk ac o s do no explain ei he he amoun o CCA cases o he inc easing inci-
dence. This scena io aises conce n abou he impac o unde explo ed ac o s beyond hese
e iologies [2].
Non-alcoholic a y li e disease (NAFLD) is ecognized as an eme ging isk ac o
o ch onic li e disease, ci hosis and li e cance , especially hepa ocellula ca cinoma
(HCC) [
3
]. I is sugges ed ha NAFLD- ela ed HCC may ca y a wo se p ognosis and
poo e esponse o sys emic ea men s compa ed o o he e iologies [
4
]. The e ha e
been e o s owa d he cha ac e iza ion o pa ien s wi h me abolic diso de s who a e
a isk o de eloping ci hosis and p ima y li e cance . Recen ly, a mo e p ac ical and
inclusi e app oach sugges ed he nomencla u e o me abolic-associa ed a y li e disease
(MAFLD) [
5
]. The MAFLD de ini ion highligh s he impo ance o o e weigh /obesi y and
ype 2 diabe es melli us as key addi ional ac o s o a s ea o ic li e . Li e disease ela ed o
me abolic ac o s cha ac e izes he hepa ic mani es a ion o a mul isys em diso de , which
is he e ogeneous in i s cou se and ou comes.
Recen esea ch highligh s he ole o oxida i e s ess and lipo oxici y in he p og es-
sion o li e disease and li e a deposi ion [
6
]. Hype insulinemia and insulin esis ance,
which occu in he con ex o DM and obesi y, a e associa ed wi h bo h a deposi ion and
malignan cell ans o ma ion [
7
]. Insulin is ound in bile and s imula es cell p oli e a ion
o choles a ic cells, which may p omo e cholangioca cinogeneis [
6
]. Lep in, a ho mone
sec e ed by adipous issue, may play a ole in cholangiocy es’ ans o ma ion, g ow h and
mig a ion [
8
]. The e o e, a biological backg ound suppo s a po en ial associa ion be ween
me abolic diso de s and CCA.
Concomi an inc eases in he incidence o me abolic diso de s and CCA may indica e a
causal ela ionship be ween hese diseases. Howe e , he e is a pauci y o da a suppo ing
his assump ion. Conside ing he po en ial ole o me abolic dys unc ion in he ca cinogen-
esis o CCA and i s impac on p ognosis, he aim o his s udy is o desc ibe he p e alence
o clinical me abolic diso de s (such as diabe es and o e weigh /obesi y) in a coho o
pa ien s wi h CCA and o epo he clinical ea u es and ou comes o pa ien s wi h CCA
and me abolic-associa ed backg ound.
2. Ma e ials and Me hods
2.1. S udy Design and Pa icipan s
We e alua ed a e ospec i e coho o pa ien s diagnosed wi h CCA om Oc obe
2013 o Janua y 2021 ea ed a “Ins i u o do Cance do Es ado de Sao Paulo” (B azil). All
pa ien s included in his s udy had con i med his ologic diagnosis o CCA ob ained h ough
pe cu aneous biopsy, ine-needle aspi a ion o su gical esec ion. Clinical cha ac e is ics,
pas medical his o y, unde lying li e disease, ea men s and ou comes we e collec ed
om he medical eco ds. Since he his ological classi ica ion was signi ican ly modi ied in
Cance s 2022,14, 3483 3 o 14
he WHO Classi ica ion o Tumo s in 2019 [
9
], a his ological e iew as well as assessmen
o new immunohis ochemical ma ke s a e unde s udy and will be epo ed a e wa d.
Pa ien s we e u he di ided in o wo g oups: (g oup 1) pa ien s wi h a pas medical
his o y (con i med o sel - epo ed) o diabe es melli us (DM) o /and body mass index
(BMI) o
≥
25 kg/m
2
and no pas his o y o li e disease, and (g oup 2) pa ien s wi hou any
o he ea u es men ioned. Among he pa ien s in g oup 1 (“me abolic diso de g oup”),
he imaging ea u es, ea men and ou comes we e de ailed. The s udy was app o ed by
he ins i u ional e hics commi ee (p o ocol numbe 3.807.496).
2.2. Managemen and T ea men P o ocol
Acco ding o he local p o ocol, all pa ien s e e ed o ou ins i u ion ha e CCA
diagnosis con i med by eassessmen o he ex e nal issue sample o newly ob ained
samples. Baseline e alua ion consis s o pe o mance s a us assessmen using he Eas e n
Coope a i e Oncology g oup (ECOG) scale, pas medical his o y, como bidi ies, li e
unc ion es s and gene al labo a o y pa ame e s. Radiological s udies a e pe o med
o assess he loco- egional o dis an sp ead, s aging and esec abili y. Imaging consis s
o ches , abdomen and pel is compu ed omog aphy (CT) scans. Whene e equi ed,
li e magne ic esonance imaging (MRI) o cholangiopanc ea og aphy a e pe o med.
Se um umo ma ke s, such as ca bohyd a e an igen (CA) 19–9 and ca cinoemb yonic
an igen (CEA), a e ou inely collec ed a he baseline assessmen and pe iodically du ing
he ea men and ollow-up.
T ea men s a e p o ided acco ding o he local p o ocol, which is in line wi h he main
guidelines adop ed globally [
10
]. B ie ly, he ea men s a egy a ies o each ype o CCA
depending on i s si e o o igin. Pa ien s wi h local disease who a e po en ial candida es o
esec ions a e usually discussed in weekly mul idisciplina y umo boa ds o su ge y indi-
ca ion. A e esec ion, no adju an ea men is ou inely o e ed, and pa ien s a e ollowed
wi h imaging assessmen e e y 6 mon hs. Pa ien s wi h locally ad anced/un esec able
disease and hose who ha e dis an me as asis a e conside ed o sys emic ea men . Pa-
ien s wi h ECOG pe o mance s a us o 0–2 wi h no o gan dys unc ions a e sui able o
i s -line chemo he apy wi h cispla in-gemci abine combina ion, acco ding o he ABC-02
ial [
11
]. Al e na i ely, some pa ien s may s a gemci abine mono he apy o o he egi-
mens (gemci abine-oxalipla in o luo opy imidine-oxalipla in) as i s -line ea men a
he physician’s disc e ion. A e p og ession o i s -line he apy, pa ien s who a e i o
subsequen ea men a e o en conside ed o ecei ing 5- luo ou acil-based egimens as
a second-line ea men . Whene e equi ed, candida es o sys emic ea men ecei e
bilia y d ainage.
2.3. Pa ien s wi h a Backg ound o Me abolic Diso de s
Pa ien s wi h a pas medical his o y o DM and/o BMI o
≥
25 kg/m
2
we e selec ed
om he o al coho and g ouped as a “me abolic diso de ” g oup. Clinical cha ac e is ics
and ou comes o his speci ic g oup we e analyzed. Baseline images we e also e alua ed
in o de o assess he p e alence o s ea osis. The manual mean li e a enua ion was
measu ed in Houns ield uni s (HU) by using a simple and p e iously alida ed echnique,
which consis s o he placemen o a ROI o e a ep esen a i e pa enchymal po ion o he
igh hepa ic lobe. The c i e ia used o de ining s ea osis we e li e a enua ion
≤40 HU
on unenhanced CT images. Al hough many c i e ia ha e been p e iously used o de e mine
li e s ea osis wi h a iable sensi i i y and speci ici y, i is sugges ed ha a li e a enua ion
alue
≤
40 HU ep esen s he mos accu a e c i e ion o de ec ing mode a e- o-se e e
disease [12,13].
2.4. S a is ical Analysis
Desc ip i e me hods we e used o analyze he incidence o isk ac o s in he o al
coho and o epo baseline and demog aphic ea u es. Con inuous a iables we e
exp essed as mean, median, anges o in e qua ile anges (IQR). Ca ego ical a iables
Cance s 2022,14, 3483 4 o 14
we e exp essed as equency. Compa isons be ween he g oup o in e es (“me abolic
diso de g oup”) and he g oup o pa ien s wi h no backg ound o me abolic diso de s
we e pe o med. Ca ego ical a iables we e compa ed using he χ2- es o Fische ’s exac
es when app op ia e. Con inuous a iables we e compa ed using S uden ’s - es . O e all
su i al (OS) was es ima ed using he Kaplan–Meie me hod, and cu es we e compa ed
by using log- ank. Fo he analysis including he whole coho , a Cox eg ession model,
including a iables ha showed signi icance in he uni a ia e analysis, was pe o med
o e alua e he independen p ognos ic ac o s and calcula e he haza d a ios (HR) and
95% con idence in e als (CI). Fo ime- o-e en analysis including only pa ien s submi ed
o su ge y ( ecu ence- ee su i al and o e all su i al om su ge y), a Cox eg ession
model was pe o med, including a iables associa ed wi h ou comes in pa ien s wi h CCA
submi ed o esec ion [
14
–
16
]: p ima y si e (in ahepa ic, pe ihila o dis al), ascula
in asion (yes o no), s a us o esec ion ma gin (R0, R1 o R2), baseline CA 19.9 (< o
≥150 U/mL
), nodal disease (N1 o N0), numbe o nodules (uni- o mul inodula ) and
ECOG-PS (0, 1 and
≥
2). A p< 0.05 was conside ed signi ican . Da a we e e alua ed using
he STATA so wa e e sion 15.0.
3. Resul s
3.1. Baseline Cha ac e is ics
F om Oc obe 2013 o Janua y 2021, 122 pa ien s wi h con i med diagnosis o CCA
we e included in he p esen analysis. The median age was 62 yea s (IQR 55–67), 72
(59%) we e emale, and he p edominan p ima y si e was in ahepa ic CCA (n= 48; 39%)
ollowed by dis al CCA (n= 41, 33.6%) and pe ihila CCA (n= 30, 24.6%). His o y o alcohol
consump ion was epo ed by 20 (16.4%) pa ien s and smoking by 54 (44.3%). Rega ding
me abolic ac o s, he median BMI was 23 kg/m
2
[IQR 20–26], 23 (18.9%) pa ien s had BMI
25–30 kg/m
2
, 13 (10.7%) pa ien s had BMI
≥
30 kg/m
2
, 24 (19.7%) pa ien s had DM, and
8 (6.6%) pa ien s had bo h DM and BMI
≥
25 kg/m
2
. The o he isk ac o s ound we e
PSC in wo (1.7%) pa ien s, ci hosis in one (0.8%), i al hepa i is in ou (3.2%) and human
immunode iciency i us (HIV) in ec ion in wo (1.6%) pa ien s. The majo i y o pa ien s
had a pe o mance s a us o 0–1 (n= 84; 68.9%), and 65 (53.8%) had me as a ic/un esec able
disease a diagnosis. Median gamma-glu amyl ans e ase (GGT = 423 U/L) and alkaline
phospha es (ALP = 272 U/L) we e highe han he uppe limi o he e e ence ange
(
30 U/L
and 150 U/L, espec i ely), while he o he median labo a o y pa ame e s we e
wi hin he no mal ange o only sligh ly al e ed.
Rega ding pa ien s wi h DM, 23 pa ien s we e classi ied as ha ing ype II DM, and
1 pa ien
had ype I DM. The median ime om DM diagnosis o CCA diagnosis was
6 yea s
(IQR: 2.5–12.2 yea s), and all pa ien s epo ed egula ollow-up since diagnosis.
A CCA diagnosis, 18 (75%) pa ien s we e using me o min, 9 (37.5%) pa ien s we e using
glicazide, 4 (16.7%) we e using glibenclamide, and 3 pa ien s we e using dapagli ozin
(12.5%). Ele en (45.8%) pa ien s epo ed ou ine use o insulin, while h ee (12.5%) pa ien s
epo ed p e ious use be o e CCA diagnosis. Mo e han one an i-DM d ug was equi ed
o
15 (62.5%)
pa ien s, and 2 (8.3%) pa ien s epo ed DM con ol wi h die a y and li es yle
habi s wi hou cu en medica ion. Baseline hemoglobin A1C (wi hin 3 mon hs o CCA
diagnosis) was a ailable o 19 (79.1%) pa ien s, wi h a median o 7.3% (IQR, 6.41–8.24%).
In he non-me abolic-diso de g oup, 17 (24.3%) pa ien s had a ailable hemoglobin A1C
wi h a median o 5.3 (IQR 5.1–5.6).
Rega ding he he apeu ic s a egy, 29 (23.8%) pa ien s we e ea ed wi h up on
cu a i e-in en esec ion. Chemo he apy was deli e ed o 104 pa ien s, gi en ha cispla in-
gemci abine was he mos used egimen (n= 98; 94.2%) ollowed by gemci abine-oxalipla in
(n= 3; 2.9%), gemci abine mono he apy (n= 2; 1.9%) and mFLOX (n= 1; 0.9%). Bilia y
d ainage was equi ed in 19 (15.5%) and s en ing in 66 (54.1%) pa ien s. Table 1shows he
baseline cha ac e is ics o he en i e coho and subg oups.
Cance s 2022,14, 3483 5 o 14
Table 1.
Baseline ea u es o he o al coho and subg oups acco ding o he coexis ence o me abolic
diso de s.
Va iables To al Me abolic Diso de
G oup
Non-Me abolic-Diso de
G oup pValue
n(%) 122 52 (42.6%) 70 (57.4%)
Median age, yea s (IQR) 62 (55–67) 64 (56–68.5) 59 (53–64) 0.199
Gende 0.048
Woman, n(%) 72 (59.0%) 36 (69.2%) 36 (51.4%)
Man, n(%) 50 (40.9%) 16 (30.8%) 34 (48.6%)
Condi ions
Weigh , kilog ams (IQR) 59 (51.7–68) 67.5 (59.9–79) 55 (48–59) <0.0001
Heigh , cen ime e s (IQR) 159 (153–166) 157.5 (152.5–164.5) 160 (154–168) 0.1991
BMI, kg/m2(IQR) 23 (20–26) 26 (24.5–29.5) 21 (18–23) <0.0001
Smoking, n(%) 54 (44.3%) 13 (25%) 41 (58.6%) <0.0001
Alcohol, n(%) 20 (16.4%) 6 (11.6%) 14 (20%) 0.212
PSC, n(%) 2 (1.7%) 0 (0%) 2 (2.9%) 0.219
Choleli hiasis, n(%) 11 (9.0%) 4 (7.7%) 7 (10%) 0.354
Ci hosis, n(%) 1 (0.8%) 0 (0%) 1 (1.4%) 0.387
Vi al hepa i is 4 (3.2%) 2 (3.8%) 2 (2.9%) 0.742
HIV 2 (1.6%) 0 (0%) 2 (2.9%) 0.219
P ima y si e 0.691
In ahepa ic 48 (39.3%) 18 (34.6%) 30 (42.9%)
Pe ihila 30 (24.6%) 13 (25%) 17 (24.3%)
Dis al 41 (33.6%) 19 (36.5%) 22 (31.4%)
Unde ined 3 (2.5%) 2 (3.9%) 1 (1.4%)
Pe o mance s a us 0.636
0–1, n(%) 84 (68.9) 37 (71.1%) 47 (67.1%)
2–4, n(%) 38 (31.1%) 15 (28.9%) 23 (32.9%)
Numbe o li e nodules 0.124
Unicen ic, n(%) 73 (59.9%) 37 (71.2%) 36 (51.4%)
Mul icen ic, n(%) 44 (36.1%) 13 (25%) 31 (44.3%)
Non-Applicable, n(%) 5 (4.1%) 2 (3.9%) 3 (4.3%)
Vascula in asion, n(%) 44 (36.1%) 17 (32.7%) 27 (38.6%) 0.576
Clinical posi i e node, n(%) 55 (45.1%) 26 (50%) 29 (41.4%) 0.178
Dis an me as asis, n(%) 61 (50%) 25 (48.1%) 36 (51.4%) 0.595
Tumo s a us (pT and/o cT) 0.100
T1, n(%) 9 (7.4%) 7 (13.5%) 2 (2.9%)
T2, n(%) 39 (32.0%) 21 (40.4%) 18 (25.7%)
T3, n(%) 42 (34.4%) 14 (26.9%) 28 (40%)
T4, n(%) 25 (20.4%) 7 (13.5%) 18 (25.7%)
Tx, n(%) 7 (5.7%) 3 (5.8%) 4 (5.7%)
Nodal s a us (pN o cN) 0.541
Cance s 2022,14, 3483 6 o 14
Table 1. Con .
Va iables To al Me abolic Diso de
G oup
Non-Me abolic-Diso de
G oup pValue
N0, n(%) 51 (41.8%) 18 (34.6%) 33 (47.1%)
N1, n(%) 53 (43.4%) 24 (46.2%) 29 (41.4%)
N2, n(%) 4 (3.3%) 3 (5.8%) 1 (1.4%)
Nx, n(%) 14 (11.5%) 7 (13.5%) 7 (10%)
Me as asis (cM and/o pM) 0.768
M0, n(%) 57 (46.7%) 27 (51.9%) 30 (42.9%)
M1, n(%) 65 (53.8%) 25 (48.1%) 40 (57.1%)
G ow h pa e n 0.174
Mass Fo ming, n(%) 72 (59.0%) 27 (51.9%) 45 (64.3%)
Pe iduc al in il a ing, n(%) 23 (18.9%) 10 (19.2%) 13 (18.6%)
In aduc ula g ow h, n(%) 6 (4.9%) 1 (1.9%) 5 (7.1%)
No a ailable, n(%) 22 (18.0%) 15 (28.5%) 7 (10.0%)
Labo a o y
AST, U/L median (IQR) 44.5 (29–76) 43.5 (24.7–57.2) 46 (32–75.5) 0.848
ALT, U/L median (IQR) 44 (30–68) 45 (29–74) 44.5 (30–76.8) 0.202
ALP, U/L median (IQR) 272 (141–569) 226 (143–429) 289 (134.5–588) 0.189
GGT, U/L median (IQR) 423 (158–752) 413 (132.7–630.5) 423 (165–793) 0.098
Bili ubin, mg/dL median (IQR) 1.03 (0.45–4.50) 0.98 (0.43–4.01) 1.25 (0.45–4.48) 0.707
Choles e ol, mg/dL median (IQR) 148 (132–201) 201 (181–213) 141 (130–143.5) 0.404
LDL, mg/dL median (IQR) 93 (94–129) 127 (109–140) 87 (74–91) 0.383
TG mg/dL, median (IQR) 118.5 (92–188) 132 (112–199) 92 (71–114) 0.777
AFP, ng/mL median (IQR) 3.1 (2–5.7) 3 (2.2–4.0) 3.35 (2.05–5.37) 0.157
CEA, ng/L median (IQR) 3.61 (2.08–14.5) 3.62 (2.05–12.87) 3.65 (2.1–16.87) 0.223
CA19.9, ng/L median (IQR) 145 (35.6–614) 112.45 (38.82–461.75) 151.85 (26.64–675.2) 0.855
Glucose, mg/dL median (IQR) 104 (100–106) 111 (98.5–125) 95 (83–107.5) 0.542
Albumin, g/dL median (IQR) 3.6 (3.3–4.1) 3.5 (3.1–4.1) 3.7 (3.3–4.1) 0.164
Neu ophils, /mm3median (IQR) 6620 (5800–7400) 6600 (5800–7400) 6700 (5700–7405) 0.621
Lymphocy es, /mm3median (IQR) 2100 (1400–2950) 2300 (1500–3020) 1900 (1395–2950) 0.701
Hemoglobin, g/dL median (IQR) 11.9 (10.8–12.9) 12.05 (11.0–13.07) 11.6 (10.4–12.7) 0.456
Pla ele s, 103median (IQR) 270 (201–351) 279 (185–347) 266 (212–354) 0.173
INR, median (IQR) 1.15 (1.05–1.26) 1.13 (1.05–1.24) 1.16 (1.05–1.28) 0.576
The apeu ic managemen
Cu a i e in en esec ion, n(%) 29 (23.8%) 15 (28.8%) 14 (20%) 0.256
Recu ence a e su ge y, n(%) 21 (17.2%) 8 (15.4%) 13 (18.6%) 0.039
Chemo he apy, n(%) 104 (85.2%) 45 (76.9%) 59 (84.3%) 0.256
Bes suppo i e ca e, n(%) 11 (9%) 1 (1.9%) 10 (14.3%) 0.018
Bilia y s en , n(%) 66 (54.1%) 25 (48.1%) 41 (58.6%) 0.25
Bilia y d ainage, n(%) 19 (15.5%) 9 (17.3%) 10 (14.3%) 0.328
IQR: in e qua ile ange; BMI: body mass index; PSC: P ima y Scle osing cholangi is; HIV: human de iciency
i us; AST: aspa a e amino ans e ase; ALT: alanine amino ans e ase; ALP: alkaline phospha ase; GGT: gamma-
glu amyl ans e ase; Bili: bili ubin; LDL: low-densi y lipop o ein; TG: iglyce ides; AFP: alpha- e op o ein; CEA:
ca cinoemb ionic an igen; INR: in e na ional no malized a io; U: uni ; ng: nanog ams; L: li e ; dL: decili e ; mg:
millig ams.
Cance s 2022,14, 3483 7 o 14
3.2. Compa ison be ween G oups Acco ding o Me abolic Fac o s
The g oup o pa ien s wi h me abolic ac o s (“me abolic diso de g oup”) accoun ed
o 52 (42.6%) pa ien s, while he o he 70 (57.4%) pa ien s ep esen ed he “non-me abolic
g oup”. As expec ed, he me abolic g oup had highe median weigh (67.5 s. 55.0 kg;
p< 0.0001
) and highe median BMI (26.0 s. 21.0 kg/m
2
,p< 0.0001). Mo eo e , he
me abolic g oup had a highe p e alence o women (69.2 s. 36%; p= 0.048) and non-
smoke s (75% s. 41.4%; p< 0.0001).
On he o he hand, he e was no di e ence ega ding p ognos ic ac o s, such as
ECOG-PS, me as a ic disease, posi i e lymph node, p ima y si e and choice o he apeu ic
s a egies, be ween “me abolic diso de ” and “non-me abolic-diso de ” g oups. (Table 1).
The median li e a enua ion in he “me abolic diso de g oup” was 56 UH (IQR:
46.2–56.2). Only one pa ien in he “me abolic diso de g oup” me he adiologic c i e ia
o hepa ic s ea osis, wi h a li e a enua ion o 36 UH.
3.3. Clinical Ou comes and O e all Su i al
The median OS o he “me abolic diso de ” g oup was 15.5 mon hs (95% CI
10.9–33.9
)
compa ed o 11.5 mon hs o he “non-me abolic-diso de ” g oup (95% CI 8.4–16.5; uni-
a ia e p= 0.048). In he uni a ia e analysis, o he a iables we e associa ed wi h be e
OS: CA19.9 < 150 U/mL (p= 0.0003); diabe es melli us (p= 0.04); ECOG-PS 0 (p< 0.001);
absence o me as asis (p= 0.0012); absence o ascula in asion (p= 0.001); and up on
esec ion (p< 0.001). In he mul i a ia e analysis, ECOG-PS (p< 0.0001), su ge y (p= 0.018)
and absence o ascula in asion (p= 0.048) we e independen ly associa ed wi h su i al,
while he e was no signi ican su i al di e ence be ween he “me abolic diso de ” and
“non-me abolic-diso de ” g oups in he mul i a ia e analysis (adjus ed HR: 1.09; 95% CI:
0.62–1.94). (Table 2and Figu e 1).
Table 2. Su i al by subg oups and uni- and mul i a ia e analysis.
Subg oups (n)nMedian O e all
Su i al (95% CI) Uni a ia e pHR (95% CI) Mul i a ia e,
pValue
Man 50 11.5 (8.6–20.6) 0.88
Woman 72 14.6 (9.4–17.4)
Obesi y 13 15.5 (5.8–NR) 0.77
No obesi y 109 13.4 (9.6–17.3)
Diabe es 24 22.8 (5.8–38.2) 0.04 0.54 (0.25–1.15), p= 0.111
No diabe es 98 11.9 (9.4–15.5)
Smoking 54 10.1 (7.4–15.3) 0.17
No smoking 68 16.5 (11.1–21.8)
Alcohol 20 16.2 (5.7–22.4) 0.93
No alcohol 102 14.0 (9.6–17.3)
Unde ined p ima y si e 3 3.5 (3.5–NR) 0.539
Dis al 41 20.6 (14.8–25.3)
In ahepa ic 48 10.5 (7.4–14.6)
Pe ihila 30 8.9 (5.4–26.7)
PS0 42 16.8 (14.2–29.2) <0.001 1.86 (1.43–2.41), p< 0.0001
Cance s 2022,14, 3483 8 o 14
Table 2. Con .
Subg oups (n)nMedian O e all
Su i al (95% CI) Uni a ia e pHR (95% CI) Mul i a ia e,
pValue
PS1 42 13.4 (9.4–20.6)
PS2 25 11.1 (5.5–22.0)
PS3 12 3.0 (2.1–5.4)
PS4 1 NR (NR–NR)
Vascula in asion 44 5.9 (4.5–11.4) 0.0012 1.61 (1.01–2.58), p= 0.048
No ascula in asion 74 16.5 (11.9–22.4)
Node posi i e 55 8.9 (5.8–15.3) 0.14
Node nega i e 62 16.5 (10.9–26.9)
Me as asis 61 8.9 (5.4–14.0) 0.001 1.47 (0.91–2.41), p= 0.118
No me as asis 59 21.8 (13.4–26.9)
Bilia y s en 66 13.4 (7.2–17.3) 0.07
No bilia y s en 56 15.5 (9.6–25.3)
Resec ion 29 34.2 (22.0–43.5) <0.0001 0.42 (0.21–0.86), p= 0.018
No esec ion 93 10.5 (7.5–14.0)
Family his o y o neoplasia 50 20.7 (11.9–25.2) 0.11
No amily his o y o neoplasia 67 11.5 (7.7–16.2)
CA19.9 < 150 U/mL 64 17.5 (11.5–22.4) 0.0003 1.47 (0.91–2.36), p= 0.112
CA19.9 ≥150 U/mL 58 8.1 (5.7–14.3)
Me abolic diso de g oup 52 15.5 (10.9–33.9) 0.048 1.09 (0.62–1.94), p= 0.745
Non-me abolic-diso de g oup
70 11.5 (8.4–16.5)
PS: pe o mance s a us; HR: Haza d a io; CI: con idence in e al; NR: no - eached; U/mL: Uni /millili e .
Cance s 2022, 14, x 7 o 13
diso de ” g oups in he mul i a ia e analysis (adjus ed HR: 1.09; 95%CI: 0.62–1.94). (Table 2
and Figu e 1).
Table 2. Su i al by subg oups and uni- and mul i a ia e analysis.
Subg oups (n) n Median O e all Su i al
(95%CI) Uni a ia e p HR (95%CI) Mul i a ia e, p
Value
Man 50 11.5 (8.6–20.6) 0.88
Woman 72 14.6 (9.4–17.4)
Obesi y 13 15.5 (5.8–NR) 0.77
No obesi y 109 13.4 (9.6–17.3)
Diabe es 24 22.8 (5.8–38.2) 0.04 0.54 (0.25–1.15), p = 0.111
No diabe es 98 11.9 (9.4–15.5)
Smoking 54 10.1 (7.4–15.3) 0.17
No smoking 68 16.5 (11.1–21.8)
Alcohol 20 16.2 (5.7–22.4) 0.93
No alcohol 102 14.0 (9.6–17.3)
Unde ined p ima y si e 3 3.5 (3.5–NR) 0.539
Dis al 41 20.6 (14.8–25.3)
In ahepa ic 48 10.5 (7.4–14.6)
Pe ihila 30 8.9 (5.4–26.7)
PS 0 42 16.8 (14.2–29.2) <0.001 1.86 (1.43–2.41), p < 0.0001
PS1 42 13.4 (9.4–20.6)
PS2 25 11.1 (5.5–22.0)
PS3 12 3.0 (2.1–5.4)
PS4 1 NR (NR–NR)
Vascula in asion 44 5.9 (4.5–11.4) 0.0012 1.61 (1.01–2.58), p = 0.048
No ascula in asion 74 16.5 (11.9–22.4)
Node posi i e 55 8.9 (5.8–15.3) 0.14
Node nega i e 62 16.5 (10.9–26.9)
Me as asis 61 8.9 (5.4–14.0) 0.001 1.47 (0.91–2.41), pp = 0.118
No me as asis 59 21.8 (13.4–26.9)
Bilia y s en 66 13.4 (7.2–17.3) 0.07
No bilia y s en 56 15.5 (9.6–25.3)
Resec ion 29 34.2 (22.0–43.5) <0.0001 0.42 (0.21–0.86), p = 0.018
No esec ion 93 10.5 (7.5–14.0)
Family his o y o neoplasia 50 20.7 (11.9–25.2) 0.11
No amily his o y o neoplasia 67 11.5 (7.7–16.2)
CA19.9 < 150 U/mL 64 17.5 (11.5–22.4) 0.0003 1.47 (0.91–2.36), p = 0.112
CA19.9 ≥ 150 U/mL 58 8.1 (5.7–14.3)
Me abolic diso de g oup 52 15.5 (10.9–33.9) 0.048 1.09 (0.62–1.94), p = 0.745
Non-me abolic-diso de g oup 70 11.5 (8.4–16.5)
PS: pe o mance s a us; HR: Haza d a io; CI: con idence in e al; NR: no - eached; U/mL: Uni /millili e .
52 15 5 1 0G oup = Me abolic
70 19 2 1 0G oup = No me abolic
Numbe a isk
020 40 60 80
Time (mon hs)
G oup = No me abolic G oup = Me abolic
51 15 5 1 0G oup = Me abolic
60 17 2 1 0G oup = No me abolic
Numbe a isk
020 40 60 80
Time (mon hs)
G oup = No me abolic G oup = Me abolic
Figu e 1.
Kaplan–Meie cu es showing: (
Le
) o e all su i al acco ding o subg oup wi h me abolic
diso de s s. no me abolic diso de s. The median o e all su i al o he “me abolic diso de ” g oup
was 15.5 mon hs (95% CI 10.9–33.9), and he median OS o he “non-me abolic-diso de ” g oup
was 11.5 mon hs (95% CI 8.4–16.5), adjus ed HR: 1.09 (0.62–1.94), p= 0.745. (
Righ
) o e all su i al
acco ding o subg oup wi h me abolic diso de s s. no me abolic diso de s, excluding pa ien s
managed wi h bes suppo i e ca e. The median o e all su i al o he “me abolic diso de ” g oup
was 15.5 mon hs (95% CI 11.4–33.9), and he median OS o he “non-me abolic-diso de ” g oup was
14.0 mon hs (95% CI 9.4–17.3), adjus ed HR: 1.32 (0.74–2.35), p= 0.351.
The e was no co ela ion be ween he g oup (me abolic s. non-me abolic) and he
umo si e (
χ2
= 1.46; p= 0.691). In e es ingly, he median o e all su i al o pa ien s
wi h in ahepa ic cholangioca cinoma and in he me abolic g oup (n= 30) was 7.7 mon hs
(95% CI 3.1–11.9), and o pa ien s wi h in ahepa ic cholangioca cinoma and in he “non-
Cance s 2022,14, 3483 9 o 14
me abolic” g oup (n= 18), i was 15.5 mon hs (95% CI 8.2—no eached) (uni a ia e
p= 0.006
). Howe e , a e pe o ming a Cox eg ession model including only pa ien s wi h
in ahepa ic cholangioca cinoma, he me abolic g oup was no associa ed wi h p ognosis
(HR = 0.43, 95% CI 0.13–1.42).
3.4. T ea men S a egies and Clinical Ou comes
Rega ding pa ien s who we e submi ed o up on su ge y (n= 29), 21 (75%) p e-
sen ed disease ecu ence wi h a median ime o ecu ence o 14 mon hs (95% CI 8.2–19.8).
Pa ien s in he “me abolic diso de ” g oup (n= 8) had signi ican ly longe elapse- ee
su i al (15.6 mon hs, 95% CI 8.2–24.6) compa ed o pa ien s in he “non-me abolic g oup”
(13.5 mon hs, 95% CI 3.5–19.8, mul i a ia e p= 0.02). Acco dingly, he median o e all su -
i al o he me abolic g oup a e esec ion was signi ican ly highe han he non-me abolic
g oup (43.4 [95% CI 33.9-NR] s. 21.8 mon hs [95% CI 8.6–26.9]; adjus ed HR = 0.23,
95% CI
0.06–0.86; p= 0.029) (Figu e 2).
Cance s 2022, 14, x 8 o 13
Figu e 1. Kaplan–Meie cu es showing: (Le ) o e all su i al acco ding o subg oup wi h me abolic
diso de s s. no me abolic diso de s. The median o e all su i al o he “me abolic diso de ” g oup
was 15.5 mon hs (95%CI 10.9–33.9), and he median OS o he “non-me abolic-diso de ” g oup was
11.5 mon hs (95%CI 8.4–16.5), adjus ed HR: 1.09 (0.62–1.94), p = 0.745. (Righ ) o e all su i al
acco ding o subg oup wi h me abolic diso de s s. no me abolic diso de s, excluding pa ien s
managed wi h bes suppo i e ca e. The median o e all su i al o he “me abolic diso de ” g oup
was 15.5 mon hs (95%CI 11.4–33.9), and he median OS o he “non-me abolic-diso de ” g oup was
14.0 mon hs (95%CI 9.4–17.3), adjus ed HR: 1.32 (0.74–2.35), p = 0.351.
The e was no co ela ion be ween he g oup (me abolic s. non-me abolic) and he
umo si e (χ2 = 1.46; p = 0.691). In e es ingly, he median o e all su i al o pa ien s wi h
in ahepa ic cholangioca cinoma and in he me abolic g oup (n = 30) was 7.7 mon hs (95%
CI 3.1–11.9), and o pa ien s wi h in ahepa ic cholangioca cinoma and in he “non-
me abolic” g oup (n = 18), i was 15.5 mon hs (95%CI 8.2—no eached) (uni a ia e p =
0.006). Howe e , a e pe o ming a Cox eg ession model including only pa ien s wi h
in ahepa ic cholangioca cinoma, he me abolic g oup was no associa ed wi h p ognosis
(HR = 0.43, 95% CI 0.13–1.42).
3.4. T ea men S a egies and Clinical Ou comes
Rega ding pa ien s who we e submi ed o up on su ge y (n = 29), 21 (75%)
p esen ed disease ecu ence wi h a median ime o ecu ence o 14 mon hs (95%CI 8.2–
19.8). Pa ien s in he “me abolic diso de ” g oup (n = 8) had signi ican ly longe elapse-
ee su i al (15.6 mon hs, 95%CI 8.2–24.6) compa ed o pa ien s in he “non-me abolic
g oup” (13.5 mon hs, 95%CI 3.5–19.8, mul i a ia e p = 0.02). Acco dingly, he median
o e all su i al o he me abolic g oup a e esec ion was signi ican ly highe han he
non-me abolic g oup (43.4 [95%CI 33.9-NR] s. 21.8 mon hs [95%CI 8.6–26.9]; adjus ed
HR = 0.23, 95%CI 0.06–0.86; p = 0.029) (Figu e 2).
Figu e 2. O e all su i al. (Le ) Pa ien s ea ed wi h chemo he apy: he median OS was 11.5 mon hs
(95%CI 8.4–14.8), and he e was no di e ence be ween me abolic and non-me abolic g oups (11.5 mon hs
[95%CI 5.8–15.3] s. 11.5 mon hs [95%CI 6.7–16.2]; p = 0.406). (Righ ) Pa ien s ea ed wi h su ge y: he
median o he me abolic g oup was signi ican ly be e han ha o he non-me abolic g oup (43.4 [95%CI
33.9-NR] s. 21.8 mon hs [95%CI 8.6–26.9]; HR = 0.23, 95%CI 0.06–0.86; p = 0.029.
Conside ing he pa ien s who we e ea ed wi h chemo he apy, he median OS was
11.5 mon hs (95%CI 8.4–14.8), and he e was no di e ence be ween he me abolic and non-
me abolic g oups (11.5 mon hs [95%CI 5.8–15.3] s. 11.5 mon hs [95%CI 6.7–16.2]) wi h an
adjus ed HR = 0.82 (95%CI 0.52–1.30; p = 0.406).
We obse ed a highe a e o pa ien s managed wi h bes suppo i e ca e in he non-
me abolic-diso de g oup (14.3% s. 1%, p = 0.018). The e o e, we also pe o med a
su i al analysis excluding hese pa ien s. A e excluding pa ien s managed wi h bes
suppo i e ca e, he median OS o he non-me abolic g oup (n = 60) was 14.01 mon hs
(95% CI 9.42–17.29) s. 15.47 (95% CI 11.40–33.92) o he me abolic g oup (n = 51); HR:
45 13 4 1 0G oup = Me abolic
59 16 2 1 0G oup = No me abolic
Numbe a isk
020 40 60 80
Time (mon hs)
G oup = No me abolic G oup = Me abolic
15 10 4 1 0G oup = Me abolic
14 8 1 1 0G oup = No me abolic
Numbe a isk
020 40 60 80
Time (mon hs)
G oup = No me abolic G oup = Me abolic
Figu e 2.
O e all su i al. (
Le
) Pa ien s ea ed wi h chemo he apy: he median OS was
11.5 mon hs
(95% CI 8.4–14.8), and he e was no di e ence be ween me abolic and non-me abolic
g oups (11.5 mon hs [95% CI 5.8–15.3] s. 11.5 mon hs [95% CI 6.7–16.2]; p= 0.406). (
Righ
) Pa ien s
ea ed wi h su ge y: he median o he me abolic g oup was signi ican ly be e han ha o he
non-me abolic g oup (43.4 [95% CI 33.9-NR] s. 21.8 mon hs [95% CI 8.6–26.9]; HR = 0.23, 95% CI
0.06–0.86; p= 0.029.
Conside ing he pa ien s who we e ea ed wi h chemo he apy, he median OS was
11.5 mon hs (95% CI 8.4–14.8), and he e was no di e ence be ween he me abolic and
non-me abolic g oups (11.5 mon hs [95% CI 5.8–15.3] s. 11.5 mon hs [95% CI 6.7–16.2])
wi h an adjus ed HR = 0.82 (95% CI 0.52–1.30; p= 0.406).
We obse ed a highe a e o pa ien s managed wi h bes suppo i e ca e in he
non-me abolic-diso de g oup (14.3% s. 1%, p= 0.018). The e o e, we also pe o med a
su i al analysis excluding hese pa ien s. A e excluding pa ien s managed wi h bes
suppo i e ca e, he median OS o he non-me abolic g oup (n= 60) was 14.01 mon hs
(95% CI 9.42–17.29) s. 15.47 (95% CI 11.40–33.92) o he me abolic g oup (n= 51); HR: 1.32
(95% CI 0.74–2.35), p= 0.351 Figu e 1. Acco dingly, we also pe o med a Cox eg ession
model excluding pa ien s ea ed wi h bes suppo i e ca e and obse ed ha he same
a iables emained independen ly associa ed wi h su i al (ECOG-PS: p= 0.017; su ge y:
p= 0.03; and absence o ascula in asion: p= 0.032).
De ailed esul s ega ding OS, p og ession- ee su i al and esponse a e a e shown
in Table 3.
