7/30/2019 Metabo No Mics
1/13
7/30/2019 Metabo No Mics
2/13
Inter nat ional Jour nal of Chem ical And Ph ar ma ceutical Resea rch
Volume 1, Issue 1 , July 20 12 .
Glor igin Lifesciences Pr ivate Limited.
Research Article
1
Keywords: Metabonomics, Nuclear Magnet ic Resonance, NMR, Spectroscopy, Hepatocel lular
Car cinom a, Chr on ic Liver Disea se, Diagn ost ic Tool.
Cor re spon d ing Auth or : Rake sh N Pillai, Clinical Epidem iology Reso ur ce & Tra inin g Cen tre, Med ical
College, Trivandr um 6 50 01 1 em ail address: p illai60 48 @gma il.com
Author Affiliations: 1 Cl inical Epidemiology Resource & Training Centre , Medical College,
Tr i vandr um, 2Vice Prinicpal, Medical College, Trivandrum, 3 Sr. Scientist , IISR, Papanamcode,
Trivandrum, 4 Consultant, Trivandrum
IJPCR VOL. 1 (1 ) JULY 2 01 2 w w w.ijp cr.n et 8
1H NMR BASED METABONOMICS AS A DIAGNOSTIC TOOL
TO HARACTERIZE PATIENTS WITH HEPATOCELLULAR
CARCINOMA
Rakes h N Pillai1*, Har ikum aran Nair 1, Vinay Kumar 2, Luxm i Var m a 3, Nisha Nair4
Abstract
The qu ant i ta t ive measu rem ent of dynamic mu lt i -param et r ic me tabol ic response of liv ing
systems to physiological stimuli or genetic modifications is known as Metabonomics and is
indicated to extract biochem ical inform ation with re spo nse to a biological even t. It encom pas ses
the com preh ensive and systema tic lis t ing of metabol ites and their concentrat ions in the t issue
or b iofluids. Therefore this technique sum mar izes the tem poral changes in th e wh ole or ganism,
both beneficial and adverse, through the effects of diet, l ifestyle, environment, genetics, and
ph arm aceut icals. Thus a m etabolic profile is gener ated by th e study of biofluids or tissue s an d
this data is being interp reted using chemom etr ics techniques. The technique a nalyses the ent i re
pool of metabol ite sum ma ry in a biofluid close to i ts ph enotype, thereby p rom ising a pow erful
diagn ostic tool in futur e.
In t l. J Ch em Ph a rm Re s 2 0 1 2 ;1(1 ) : 8 1 9
Clinical Research
7/30/2019 Metabo No Mics
3/13
IJPCR VOL. 1 (1 ) JULY 2 01 2 w w w.ijp cr.n et 9
1. INTRODUCTION
Metabonomics is formally defined as The
q u a n t ita t i ve m e a s u r e m e n t o f d y n a m i c
mul t i - pa r ame t r i c me t abo l i c r e sponse o f
l iving system s to physiological s t im uli orgene tic m od ifications. [1] This is the m ost
re cen t tech no logy in th e wo rld of Om ics,
which is indicated to extract biochemical
information with response to a biological
event . I t encomp asses the comp rehe nsive
and systematic l is t ing of metabol i tes and
t h e i r c o n c e n t r a t i o n s i n t h e t i s s u e o r
b i o f l u i d s . T h e r e f o r e t h i s t e c h n i q u e
summ ar i zes t he t em por a l changes i n t he
w h o l e o r g a n i s m , b o t h b e n e f i c i a l a n dadve rs e, th ro ugh th e effects of d iet, lifestyle,
e n v i r o n m e n t , g e n e t i c s , a n d
ph arm aceut icals. Thu s a m etabolic profile is
gener ated by the study of biofluids or tissues
and t h i s da t a i s be i ng i n t e r p r e t ed us i ng
c h e m o m e t r ic s t e c h n i q u e s .[ 2 ] I n s h o r t
Metabonomics analyses the ent i re pool of
me tabolite sum m ary in a biofluid close t o its
phen otype , thereby prom ising a power ful
diagnost ic tool in future. There are many
techniques u sed in Metabonom ic studies in
w h i c h N M R s p e c t r o s c o p y h a v e c e r t a i n
advantages over oth er w hich is summ arized
in Table 1 .
1.1 NMR based Metabonom ics
N uc l ea r magne t i c r e sonance ( N MR ) i s a
p o w e r f u l a n a l y t i c a l t o o l f o r t h e
c h a r a c t e r i z a t io n o f m o le c u la r s t r u c tu r e ,
qu ant itative analysis, and the exam ination ofdynam ic processes in b ioflu ids or t i s sues .
T h i s t e c h n i q u e e x p l o i t s t h e m a g n e t i c
p r o p e r t i e s o f c e r t a i n a t o m i c n u c l e i
(esp ecially atom s like Hydrogen and Carb on)
depending upon the i r chemica l s t ruc ture .
This atomic nucleus in a static magnetic field
absor bs electrom agnet ic radiation of specific
f r equency depend i ng on t he i r angu l a r
momentum,
which is recorded a s i tsresona nce pea ks on
X axis. The r eson ance pea ks takes car e of
c h e m i c a l s h i f t s w h i c h i s m a r k e d i n a
continuou s scale ran ging from 0-10 par ts per
million (PPM). The Y axis corresponds to
the inte nsity of each p eak w hich is directlyprop ort ional to the concentrat ion of those
me tabolites in the biological fluid or tissue .
Thus NMR spectroscopy wil l summarizes
t h e l i s t o f m e t a b o l i t e s a t d i f f e r e n t
c o n c e n t r a t i o n b o t h o b j e c t i v e l y a n d
subjectively.
Metabono mics based on NMR spectroscopy
have r ising ap plications in differen t are nas
of health like d iagnosis an d classification ofd i seases , to moni tor t ime-course d i sease
pr ogres sion, to iden tify new b iomar kers, to
learn pathological mechanisms, to monitor
efficiency and toxicity of diseases etc. This
i s a l s o u s e d t o g e n e r a t e m e t a b o l o m e
database s and in dru g designing.
1.2 Hepatocellular Carcinom a
Est imates f rom var ious cancer regi s t r i es
indicate that l iver cancer re mains the fi fthmost common mal ignancy in m en and the
eighth in w om en wo rldw ide. Hepa tocellular
C a r c i n o m a ( H C C ) i s t h e m o s t c o m m o n
primary l iver cancer and about 90% of al l
l iver can cer con stitutes to be HCC. The a ge
standardized incidence rates , ranges from
less than 10 cases p er 1 00,000 po pulat ions,
i n p a r ts o f No r t h Am e r i c a a n d W e s t e r n
Europ e as w ell as in Ind ia, Iran an d Iraq. The
standard m ortality rate is from 50 150 case sper 100,000 populations in parts of Africa
and Asia. In high-risk countries, HCC can
arise before the a ge of 20 years , whe reas in
coun tries at low risk, HCC is rar e b efore th e
age of 50 year s. Rates of liver can cer in m en
are typica l ly 2 to 4 t imes h igher than in
women.[3,4] .
7/30/2019 Metabo No Mics
4/13
IJPCR VOL. 1 (1 ) JULY 20 12 w w w.ijp cr.n e t 10
The i nc i dence o f p r i mar y l ive r can ce r i s
increasing in several developed countr ies ,
includ ing the United States and t he incre ase
is likely to cont inue for som e decades. The
trend is a resu lt of a cohor t effect re lated to
infection w ith hep atitis B an d C viruses , theincidence of which peaked in th e 19 50s to
19 80 s. In selected are as of some developin g
count r ies , the inc idence of pr imary l iver
cancer has decreas ed, pos sibly as a res ult of
the introdu ction of hepa titis B virus vaccine
T h e a t t r i b u t a b l e r i s k e s t i m a t e s f o r t h e
comb ined effect of this infection a ccount for
over 8 0% of liver can cer cases w orldw ide.[ 5]
T h e m a j o r e t i o l o g i e s o f h e p a t o c e l l u l a r
carcinoma are l is ted in table 2. There is a
p o s i t i v e a s s o c i a t i o n o f h e p a t o c e l l u l a r
carc inoma wi th fac tors l ike a ge , e leva ted
bod y mas s index, (esp ecially in men ) as w ell
as d iabetes m elli tus. 6 As for most types of
cancer , hepatocel lular carcinogenesis is a
mu ltistep p rocess involving differen t gene tic
altera tions th at ultimate ly lead to m alignan t
transforma tion of the hepato cytes.
1.3 Rationa le of th e study
The rat ionale for th is study wa s based on th e
concept that , populat ions at high r isk for
l iver cancer , such a s tho se w ith cirr hosis ,
have to be ident ified ea rly. Studies sh ow tha t
a r ound 20 t o 5 0% of pa t i en ts p r e sen t i ng
w i t h h e p a t o c e l l u l a r c a r c i n o m a h a d
previously undiagnosed cirrh osis . [7] The
curr ent m odalities pote nt ially available for
screening include serum alpha-fetoprotein( A F P ) a n d U l t r a s o n o g r a p h y ( U S G ) .
Moreover uncharac ter i s t ic screening m ay
l e a d t o m o r e i n v a s i v e c o n f i r m a t o r y
techn iques like liver b iopsy. Com plications
o f live r b i opsy a r e r e por t ed i n 0 .06% t o
0.32% of pat ients, and typically occur within
the first few hou rs after t he b iopsy. For this
reason, we were a iming a t new screening
strategies an d
1.4 Risk of hepa tocellular carcinom a
in chronic liver d isease
M a n y s t u d i e s s h o w i n c r e a s e d r i s k o f
hepatocel lular carcinoma in pat ients with
l iver c i r r hos i s depen ding o n t he ac t iv ity,
dura t ion and the e t io logy of the d i sease .
Clinical markers such as Alpha fetoprotein
(AFP) , Des carb oxy pr othr om bin (DCP) ,
biological var iables s uch as Hb s-Ag, an ti-HCV
a n d p l a t e l e t c o u n t s e t c a l l o w s f u r t h e r
classification of cirr hot ic patien ts w ith h igh
risk ofhepa tocellular carcinom a. Child pu gh
scor e and MELD scor e a r e ca lcu la t ed t o
m o n i t o r t h e p r o g r e s s i o n o f c i r r h o s i s .
Relative risk of liver can cer is incre ased with
coexistence of etiologies, such as hepatitis
B ( H BV) , he pa t i t i s C ( H CV) i n f ec t i on s ,
aflatoxins, [8,9] HBV, HCV infection and
alcohol or diab etes m ell itus, [10 ] ,11 HCV
i n f e c t i o n a n d l i v e r s t e a t o s i s . [ 1 0 ]
Coexistence of oth er en vironm en tal factor s
l ike a lcohol , [8 ,12 ,13] d iabetes mel l i tus ,
o b e s i t y a n d t o b a c c o a ls o i n c r e a s e s t h e
relat ive r isk of HCC development . [12,14]
Ther e is also incre ased risk w ith occult HBV
i n f e c t i o n . [ 1 5 ] I n t e r e s t i n g l y , c o f f e e
consumpt i on appea r s t o r educe t he H C C
risk.[16] In general, ther e are evidences th at
HBV an d HCV un der certain circum stances
play an a dd itiona l direct r ole in th e m olecular
pathogenesis of HCC. Aflatoxins have been
s h o w n t o i n d u c e m u t a t i o n s o f t h e p 5 3
tum our su ppr essor gene , thus p oint ing to
the contr ibut ion of an environm ental factor
to th e tum our developm ent at th e m olecular
level.
char acterization tools to find the vulne rab le
p a t i e n t s e f f e c t i v e l y w h e n p r e v e n t i v e
treatment is helpful. Metabonomics based
on 1H NMR, is very sen sitive instrum ent and
c a n b e d e v e l o p e d f o r s c r e e n i n g a n d
sur veillance of hep atocellular carcinom a. Itcan efficien t ly extra ct detai led m olecu lar
i n f o r m a t i o n o n a l a r g e n u m b e r o f
m etabolites in d iffere nt bio fluids.
7/30/2019 Metabo No Mics
5/13
IJPCR VOL. 1 (1 ) JULY 2 012 w w w.ijp cr.n e t 11
1.5 Gold stan dard in the d iagn osis of
hepa tocellular carcinom a
Standards recom m ended b y EASL (Europ ean
a s s o c i a t i o n o f s t u d i e s o n l i v e r ) i n t h e
diagnosis of liver cancer depen d o n th e size
of nodules in the liver. In the case of nodulesize not exceed ing 2 cm, biopsy of the no dule
i s r e c o m m e n d e d s i n c e t h e i m a g i n g
techn iques do not have su fficient accur acy to
dist inguish hepatocel lular carcinoma from
other ben ign or malignant condi t ions and the
AFP concentration will usually re ma in w ithin
n o r m a l v a l u e s o r b e s l i g h t l y e l e v a t e d .
Pathological confirmations are obtained by
cytology or h istology, bu t th e com binat ion of
both techniques offers the highest d iagnost ic
accuracy. In the case of no dules a bove 2 cm,
imaging techniques along with ap prop riate
s e r o l o g y c a n c o n f i d e n t l y e s t a b l i s h t h e
diagnosis without need ing confirmat ion with
a pos i t i ve b i op sy .[ 17 ] . The EASL, ha s
stressed that the biopsy of small lesions m ay
not be rel iable due to 3 r easons,
In the case of small lesions needle
placeme nt may b e d ifficul t and one
cannot be cer tain that the samp le was
drawn from the lesion.
There is a d ispar i ty between
pathologists in sort ing out dysplasia
and HCC, and this disagreement occurs
more frequently as the size of the
lesion decreases.
It may b e d ifficult to dist inguish we ll-
d i f f e r e n t i a t e d h e p a t o m a f r o m
normal l iver on b iopsy, where the
archi tectural features of l iver cel ls
such a s w idened plates m ight be lost .
the d iagnosis of l iver cancer. The pr imar y
objective was to identify and characterize
the distribution o f significant me tabolites in
s e r u m o f p a t i e n t s w i t h h e p a t o c e l lu la r
c a r c i n o m a a n d p a t i e n t s w i t h o u t
hep atocellular car cinom a, from a bas ket ofm e t a b o l o m e l i b r a r y . T h e s e c o n d a r y
ob jective wa s to der ive an o pt imal cut-off for
each m e t abo l it e w h i ch can d i ffe r en t i a t e
be tween these two cond i t ions . Since we
w er e us i ng t he p r i nc i p l e s o f d i agnos t i c
test ing in th e ana lysis, the sa m ple size was
ca lcu la t e d accor d i ng l y, u s i ng n - ma s t e r .
Keep ing sens i t iv ity of new tes t as 9 5 % ,
pre cision of pop ulat ion param eter as 1 0 % ,
a n d ( 1 - a ) a t 9 5 % , t h e n t h e s a m p le s i zerequ ired was 18 p osi t ive cases.
2. MATERIALS AND METHODS
T h i s s t u d y w a s a v e n t u r e t o a d o p t
ep idem iological pr inciples into b asic science
resea rch an d is a sole attemp t to highlight the
use of Nuclear ma gnet ic reson ance (NMR)
spe ctroscop y and Metabo nom ics pr inciple in
2.1 Sam ple Collection
The patients were selected from Gastroen-
t e r o l ogy depa r t men t , Med i ca l C o l l ege ,
Trivandrum, which is a ren owne d ter t iary
care centre for liver related ailm ents. Seru m
ana lysis using NMR spectroscop y, was car -
r ied out w ith the help and guidance from
NIST (Nationa l inst itut e of science an d tech -nology), Trivandrum, which is one among
the recognized CSIR labs in India. All sus-
pected cases of chronic liver disease con-
secu tively adm itted to Gastro en ter ology de-
pa rtm en t in Med ical College, with r esu lts of
u ltra soun d examinat ion and approp r ia te
serology tests were included in the study.
Patients not giving consent as we ll as th ose
without o r n ot w ill ing for fur th er tests wer e
excluded. For ROC ana lysis gold stan dar d
considered were ul t ra sound examinat ion
along w ith app ropriate ser ology and expert
opinions . Appr oval from hum an e th ica l
commit tee was ob tained and th e data were
col lec ted us ing Per forma af te r informed
consen t which was com pleted by ph ysician
at tending the pat ient .
Blo o d sa m p le s w e r e co lle ct e d an d
r o u t i n e b i o c h e m i c a l p a r a m e t e r s
were m easured.
7/30/2019 Metabo No Mics
6/13
IJPCR VOL. 1 (1 ) JULY 2 01 2 w w w.ijp cr.n et 12
and at tent ion. The m ost impor tant facet of
a n y f i t t i n g t e c h n i q u e i s a p p l y i n g i t
consisten tly to all spe ctra in a dataset.
F i r s t s t e p w a s t o i d e n t i f y a l i s t o f
metabolites (targeted profile) from a pool
of comp oun ds available in the Chen om x
l i b r a r y . L a t e r t h i s c o m p o u n d s e t w a s
ma tched an d qu ant i fi ed uni form ly in a l l
e x p e r i m e n t a l s p e c t r u m s . I n t h i s s t u d y
p r o f il in g w a s r e s t r ic t e d t o c o m p o u n d s
reson at ing their peaks in the re gion 0- 4.5
pp m. There are two w indows in th is region,
wh ich const itute s lipopr otein lipids (LIPO)
w i t h compl ex b r oad spec t r ums o f l i p i d
m o l e c u l e s a n d l o w m o l e c u l a r w e i g h t
molecules (LMWM) like glucose.[ 1 9]
Init ially 2 50 me tabol i tes wer e considered
from the CHENOMX libra ry an d pr ofiling
was done on 137 metabol i tes . The others
w e r e e x clu d e d b e c a u s e o f s u s p e c t e d
overlaps with water peaks. Quantification
was b ased on the intensi ty of peaks, in th e
e x p e r im e n t a l s p e c tr u m m a t c h e d w i t h
reference peaks . Prof i l ed compounds as
wel l as i ts po tent ia l concent ra t ion w ere
e x p o r t e d t o E x c e l s h e e t f o r s t a t i s t i c a l
analysis.
2.21
H NMR Spectroscop y
The serum was d i luted by D2O (Deuter ium
oxide) a t a ra t io of 1 : 4 l concent ra t ion .
500 l of samp le is transferred into a 5 m m-
outer -diameter NMR tubes an d was pr ocessed
p r o c e s s e d i n a B r u k e r s p e c t r o m e t e r
o p e r a t i n g a t 5 0 0 M H Z . F o r a c q u i r i n g
quan t i ta t i ve mea su r em en t s t he s equence
was re pea ted 4 0 time s each with a total delay
of 5.8s and an a cquisition time 1 .7s, which
g a ve r e a s o n a b le s i g n a l t o n o i s e ( S / N )
ratio.[18].
The la rge water p eaks was sup pressed by
ap plying a s econda ry irra diat ion field at i ts
resonance f requency. The procedure was
repe ated in al l sam ples un iform ly to br ing
intern al val idi ty in data pr ocessing. Phase
angle and base l ine correct ions w ere done
a n d t h e s p e c t r u m i s t r a n s f e r r e d i n t o
C h e n o m x p l a t f o r m f o r t a r g e t p r o f i l i n g .
Targeted pr ofiling involves fit t ing a se ries o f
compounds to an exper imenta l spec t rum.
Practically, targeted profiling includes the
two tasks of ident i fy ing compounds , and
f i t t i n g t h e i n d i v i d u a l c l u s t e r s o f e a c h
compou nd to th eir respect ive regions of the
spe ctru m . Fitting individua l com pou nds in a
region of the spe ctrum requ ires some care
Serology for Alfa fe toprotein and Ultra
s o n o gr a m w e r e m e a s u r e d o n a l l
p a t i e n ts . CT a n d M R I w e r e u s e d
wherever required.
Ch ild P u gh s co r e we r e r eco r de d a s a
sum mar y index in the progression of liver cirrh osis.
Th e pa t ie n ts w e r e g r o up e d, a s
h e p a t o c e llu la r c a r c in o m a a n d n o
hepatocellular carcinoma, according
to th e re sults from gold stan dards.
Th e se r um s am p le s w e re se n t fo r
N u c l e a r M a g n e t i c R e s o n a n c e
Spectroscopy.
2.3 ROC an alysis
Di agnos t i c pe r fo r m ance o f a t e s t o r
accuracy of a test to d iscr iminate d iseased
cases from non diseased is evaluated using
Receiver Operat ing Character is t ic (ROC)
curve a na lysis. [20] ROC curves can a lso b e
u s e d t o c o m p a r e t h e d i a g n o s t i c
per form ance of two or mor e labora tory or
d i a g n o s t i c t e s t s . [ 2 1 ] R O C ana l yses a r e
comm only used in me dical decision m aking,
a n d i n r e c e n t y e a r s h a v e b e e n u s e d
increasingly in machine learning and data
mining research. For a cont inuous ou tcome,
as s een in th is s tudy, differen t th res holds
(cut offs or criterion value) ar e ap plied to
pred ict class me mb ersh ips.
7/30/2019 Metabo No Mics
7/13
IJPCR VOL. 1 (1 ) JULY 2 01 2 w w w.ijp cr.n et 13
T h i s i s a n e s t i m a t e o f e a c h m e t a b o l i t e s
pr oba bility to wh ich indepen dent class t hey
be lon gs to, at t ha t cut o ff [22 ].
T o t a l 1 3 7 m e t a b o l i t e s i n t h e
m e t a b o l o m e w e r e c o n s i d e r e d a s 1 3 7indepe ndent te sts an d con tingen cy table for
each m etabolites an d th ere se nsitivity as well
as false p ositive r ates, at ea ch cut offs w ere
ca l cu l a t ed . R O C cur ve p l o t t ed and t he i r
corresponding area under the ROC (AUC)
wer e com pu ted. AUC gives the pr oba bility
that , when presen ted w ith a ran domly chosen
pat ient w ith disease and a ran domly chosen
pat ient withou t disease, the resu lts of the te st
will rank the p at ient w ith disease as havinghigher chance for d isease than the pa t ient
w i t h o u t d i s e a s e . T h e r e fo r e t h o s e
M e t a b o l i t e s w i t h h i g h e r A U C c a n b e
c o n s i d e r e d a s t h e b e s t m a r k e r s t o
d iscriminate hep atocellular carcinom a from
no ca r c i nom a . Op t i ma l cu t -o f fs f o r e ach
me tabol it e can be read d i rec t ly f rom ROC
curves.Several comb ination s of m etabolites
whe re t r ied so that , those m etabol ites wh en
compared paral lel , ( i .e . , i f any one test isp o s i t i v e i s c o n s i d e r e d a s t e s t p o s i t i v e )
improves th e sen si t ivity at minimal cost of
specificity.
be tween 41 - 50 age groups .Biochemical
par am eter s sh owing th e elevated levels of
liver en zyme s and CHILD-Pugh scor es for all
pat ients, wh ich ar e sum m arized in Table, 3.
ROC curve was plotted for each m etabolite
a t d i ffe ren t concen t ra t ions a nd AUC wascalculated (Table 4). Metabolite, which had
significant AUC (> 0.70) were selected for
furth er analysis . Out of 137 me tabol i tes 7 3
w e r e f o u n d t o h a v e AUC a b o v e 0 .7 0 a s
shown in the table. An optimal cut-off from
ROC curve was obtained and the pat ients
wer e classified into binar y group s as having,
h e p a t o c e l l u l a r c a r c i n o m a o r N o
h e p a t o c e l lu la r c a r c i n o m a . M e a s u r e o f
d i a g n o s t i c a c c u r a c y l i k e S e n s i t i v i t y ,S p e c i f i c i t y a n d L i k e l i h o o d r a t i o s f o r
hepa tocellular carcinom a w ere compu ted.
The op timal cut-off was s elected at th e po int
w her e , t he r e w as a r ea sonab l e t r ade - o f f
betwee n sen sitivity an d s pe cificity. Here the
investigator pr eferred b etter sen sitivity for
selecting cut off values , un like in t he case o f
tumor markers where specif ici ty is given
more importance. This strategy is adopteda t t h i s s t a g e k e e p i n g i n m i n d t h a t , a
scree ning tool is essen tial for th e detect ion
of ea r l y hepa t oce l l u l a r ca r c i noma f r om
cirr hot ic patients. The final m etabolom e wa s
d r a w n fr o m t h e t a b le , w h i ch h a v e h i g h
l ike l ihood pos i t ive va lues (> 3) and less
likelihood negat ive values (Table 5). These
are th e m etabolit es in th e an a lys is wh ich
have the highest l ikel ihood to d iscr iminate
hep atoce l lu lar carc inom a. Many par a l le lcombinat ions where t r i ed on the bas i s of
their l ikel ihood rat ios. Those with higher
l ikel ihood posi t ive values were pooled to
check the improvement in sens i t iv i ty . In
par a l le l comb inat ions (Table 6 ) th e t e s ts
were considered positive if any one of the
me tabol i tes in the combinat ion is posi t ive.
3. RESULTS
There wer e 68 pat ients part icipated
in th e stud y. On the bas is of resu lts from gold
s tandards , 20 pa t ients w here grouped in to
hep atocellular carcinom a, 28 as chr onic liver
disease with cirrh osis and 2 0 as appar ent lynor ma l liver. Therefore th e en t i re da tase t
obtaine d h ad, 13 7 m etabolites X 68 pa tien ts.
Mean age distribution am on g hep atoce llular
carcinom a pat ients was 5 0.21 (45.81-54.61)
and that of pat ients group ed as Chron ic liver
disease w as 5 0.79 (46.86-54.71) . Mean age
distr ibut ion among people with apparent ly
nor ma l liver wa s 40.15 (36 .81 -43.49 ). 52 .6%
of hep atocellular carcinom a pat ients wer e
7/30/2019 Metabo No Mics
8/13
IJPCR VOL. 1 (1 ) JULY 20 12 w w w.ijp cr.n e t 14
9 . Min g , L., e t a l. D o m i n a n t r o le o f
hep atitis B virus an d cofactor r ole of
aflatoxin in he pa tocarcinogen esis in
Qidon g, Chin a. Hep a to logy 36 , 121 4-
1220 ( 2002) .
1 0 . Hassan , M.M., e t a l. Risk factors forhep atocellular carcinom a: syner gism
of a lcohol w i th v i ra l hep at i t i s and
d i abe t e s me l l i t u s . H e p a t o lo g y 36 ,
1206- 1213 ( 2002) .
11 . Ohata , K., e t a l. Hepat ic stea tosis is a
r i s k f a c t o r f o r h e p a t o c e l l u l a r
carcinoma in pat ients with chronic
hep atitis C virus infection. Cancer97 ,
3036- 3043 ( 2003) .
1 2 . Mar re r o , J.A., e t a l. Alcohol, tobaccoa n d o b e s i t y a r e s y n e r g i s t i c r i s k
factor s for h ep atocellular car cinom a.
J He p a to l42 , 218-224 (2 005) .
13 . Mor gan , T.R., Mandayam, S. & Jama l,
M. M. A l coho l and hepa t oce l l u l a r
c a r c i n o m a . G a s t r o e n t e r o lo g y 1 27 ,
S87-96 (200 4) .
14 . El-Serag, H.B., Tran , T. & Everhar t, J.E.
Diabete s increases th e risk of chron ic
l iv e r d i s e a s e a n d h e p a t o c e l lu la r
c a r c i n o m a . G a s t r o e n t e r o lo g y 1 26 ,
460- 468 ( 2004) .
1 5 . Pollicin o , T., e t a l. Hepatitis B virus
m a i n t a i n s i t s p r o - o n c o g e n i c
pro per ties in the case of occult HBV
infection. Gastroenterology126 , 102 -
110 ( 2004) .
16 . Gela tt i, U., e t a l. Coffee consumption
reduces the r i sk of hepatoce l lu lar
c a r c i n o m a i n d e p e n d e n t l y o f i t s
ae t i o l ogy : a ca se - con t r o l s t udy . J
Hep a tol42 , 528-534 (2 005) .
17 . Br u ix, J., e t a l. Clinical management
o f h e p a t o c e l l u l a r c a r c i n o m a .
Conclusions of the Barcelona-200 0
E A S L c o n f e r e n c e . E u r o p e a n
Asso ciation for t he Stud y of th e Liver.
J He p a to l35 , 421-430 (2 001) .
4. REFERENCES
1 . N icho lson , J.K., Conne l ly, J., Lindon ,
J.C. & Holmes, E. Metabonomics: a
platform for studying drug toxicity
a n d g e n e f u n c t i o n . N a t Re v Dr u gDisco v 1 , 153-161 (2 002) .
2 . Nicholson , J.K., Lindon , J.C. & Holmes ,
E. Metabonomics: understanding
t he me t abo l ic r e sp onses o f liv ing
s y s t e m s t o p a t h o p h y s i o l o g i c a l
s t imu l i via m ul t ivar ia te s ta t i s t i ca l
ana l ys i s o f b i o l og i ca l N MR
spe ctroscop ic data. Xe n o b io t ic a 29 ,
1181- 1189 ( 1999) .
3 . She r man , M. & Kle in , A. AASLD s ingle -t o p i c r e s e a r c h c o n f e r e n c e o n
h e p a t o c e l l u l a r c a r c i n o m a :
Conference proceed ings.Hep a t o logy
40 , 1465 -1473 (2004) .
4 . Br u ix, J. & Sh er m an , M. Ma na ge m en t
o f h e p a t o c e l l u l a r c a r c i n o m a .
Hep a to logy 42 , 1208-1236 (2 005) .
5 . Bo s ch , F.X., Rib e s , J., Dia z , M. &
C l e r i e s , R . P r i mar y l i ve r cance r :
w o r l d w i d e i n c i d e n c e a n d t r e n d s .
Gastroe nte rology 127 , S5-S16 (20 04 ).
6 . Blu m , H.E. He p at oce llu la r ca r cin o m a
- G l oba l bu r den and r i sk f ac t o r s .
SUPLEMENT O IATREIA VOL 20 , S11 -
14 ( 2007) .
7 . Za m a n , S.N., Jo h n so n , P.J. & W illia m s ,
R. Silen t ci rrh osis in p at ients with
h e p a t o c e l l u l a r c a r c i n o m a .
Implicat ions for screening in h igh-
incidence and low-incidence areas.
Cancer65 , 1607-1610 (1990) .
8 . Yu , M.C. & Yu a n , J.M. En vir o n m e n ta l
factors and r isk for hepatocel lular
c a r c i n o m a . Gastro en tero logy 1 27 ,
S72-78 (200 4) .
7/30/2019 Metabo No Mics
9/13
IJPCR VOL. 1 (1 ) JULY 2 01 2 w w w.ijp cr.n e t 1 5
1 8. Nich olson , J.K., e t a l. Proton-nuclear-
m a g n e t i c - r e s o n a n c e s t u d i e s o f
s e r u m , p l a s m a a n d u r i n e f r o m
fast ing nor ma l and diabet ic subjects .
B io ch em J217 , 365-375 (1 984) .
19 . Ma kin en , V.P., e t a l. 1H N MRm e t a b o n o m i c s a p p r o a c h t o t h e
d i s e a s e c o n t i n u u m o f d i a b e t i c
complicat ions and p rem ature death.
Mol S yst Bio l4 , 167 (2 008) .
2 0 . Me tz , C.E. Ba s ic p r in c ip le s o f ROC
analysis. Semin N uc l Med 8 , 283 -298
( 1978) .
21 . Gr iner, P.F., Mayewski , R.J., Mushl in ,
A.I. & Greenland, P. Selection and
interp retation of diagnostic tests an d
p r o c e d u r e s . P r i n c i p l e s a n d
app l i ca t i ons . An n In te r n M e d 94 ,
557- 592 ( 1981) .2 2 . To m , F. ROC gr ap h s w it h in s ta n ce -
varying costs. Pattern Recogn. Let t .
27 , 882-891 (2 006) .
Table 2, lists out the major etiological factors of hepatocellular carcinoma worldwide, as cited in
the art icle by Blum et.al in r eference 6 .
Table 2, Major etiologies of hepa tocellular carcinom a
Tab le 1, Advan tages of NMR sp ectr osco py
Table 1, l ists s om e distinguishing feature s o f Nuclear Magnetic Resonan ce (NMR) s pectroscopyover other comm on techn iques in Metabonomic studies.
! Cirrho sis due to Chronic Hepat itis B, C and D
! Cirr ho sis du e to Toxins ( e.g., alcoh ol, to ba cco, aflato xins )
! Autoimm une hepa t i t is
! Hered itary m etabolic liver d iseases
(e.g., her ed itar y hem ochrom atosis, 1-antitrypsin deficien cy)
! States of ins ulin re sistance
Overw eight in ma les
Diabe tes me llitus
Non-alcoholic steatohepatitis (NASH)
Non-alcoholic fatty liver disease (NAFLD)
! Dire ct an alysis of com plex biological sam ples
! Non selective an d simu ltaneou s detection o f en dogen ous m etabolites
! Quan ti tat ive at minimu m of 10 M concentrat ion
! Rapid, approximately 10 minute s
! Sam ples ar e no n destructive
7/30/2019 Metabo No Mics
10/13
Table 3, shows t he biochem ical par am eter s of 68 pat ients in the s tudy who w ere groupe d as HCC, CLD and
normal l iver. Abbrevations for, AST (aspartate transferase), ALT (alkaline transferase), ALP(alkaline
pho spatase). IU/ L is the intern ational unit per l it re and m g/ dl is mill igram p er desili tre.
Tab le 4 , List of me tabolit es w ith significant AUC
IJPCR VOL. 1 (1 ) JULY 20 12 w w w.ijp cr.n e t 16
Table 3, Biochem ical pa ra me ter s of th e t otal sam ples
Cha ra cter ist ics HCC grou pn =20 No HCC grou pn =28
(Mean , 95% C.I) (Mean , 95 % C.I)
AST (IU/ L) 180 (134 .02-225 .98) 104 .07(8 0.71-12 7.43)
ALT (IU/ L) 80 .68 (70 .18 -91.19) 58 .21 (45 .49 -70.93)
ALP (IU/ L) 284 .26(1 83.43-385.10) 165 .04(1 36.04-194.03)
Bil irub in (mg/ dl) 4.068 (2.067 -6.069 ) 2.636 (1.950 -3.322 )
CHILD-Pugh 11 (10 .26 -11.74) 8.75(7 .90 -9.60 )
Ar ea 95% Confidence
Sl no un der P va lue In ter va l
curve
1 . Galactar ate 0 .8 57 0 .0 00 0 .7 65 0 .9 49
2 . Ace toace tate 0 .8 50 0 .0 00 0 .7 60 0 .9 40
3 . Methylam in e 0 .846 0 .00 0 0 .75 5 0 .9 364 . 1 ,3 -Dihydr oxyace tone 0 .8 35 0 .0 00 0 .7 41 0 .9 30
5 . 2 -Hydr oxyglu tarate 0 .835 0 .000 0 .7 34 0 .93 6
6 . S-Su lfocyste in e 0 .83 0 0 .0 00 0 .73 1 0 .9 30
7 . 3 -Hydr oxy-3 -Methylglu tar ate 0 .8 28 0 .0 00 0 .7 31 0 .9 25
8 . Bu tan on e 0 .827 0 .000 0 .72 8 0 .92 7
9 . Citr ate 0 .821 0 .0 00 0 .7 20 0.9 22
10 . Tar tr ate 0 .82 1 0 .00 0 0 .721 0 .92 1
11 . Glu tara te 0 .82 0 0 .00 0 0 .717 0 .92 3
12 . Glu tam ate 0 .81 8 0 .00 0 0 .70 5 0 .93 0
13 . Levu lina te 0 .8 18 0 .0 00 0 .71 4 0 .92 1
14 . Succin ate 0 .816 0 .000 0 .713 0 .91815 . Malate 0 .811 0 .000 0 .707 0 .91 6
16 . Suber ate 0 .810 0 .000 0 .702 0 .91 9
17 . 2 -Am inoad ipa te 0 .809 0 .000 0 .695 0 .92 4
18 . Glycin e 0 .80 8 0 .00 0 0 .70 5 0 .91 2
19 . Dim e thylam ine 0 .807 0 .000 0 .70 4 0 .91 1
20 . 2 -Oxovalera te 0 .805 0 .00 0 0 .69 9 0 .9 11
21 . Prop ionate 0 .804 0 .000 0 .69 8 0 .91 1
22 . Pyr uvate 0 .80 4 0 .00 0 0 .69 9 0 .91 0
23 . Ace ton e 0 .80 3 0 .00 0 0 .70 0 0 .90 6
24 . Glucar ate 0 .80 2 0 .00 0 0 .691 0 .91 4
25 . O-Phosphoser ine 0 .7 98 0 .000 0 .683 0 .91 3
7/30/2019 Metabo No Mics
11/13
IJPCR VOL. 1 (1 ) JULY 2 012 w w w.ijp cr.n e t 17
Ar ea 95% Con fiden ce
Sl no un der P va lue In ter va l
curve
26 . 1 ,6 -An hydr o- -d-glucose 0 .796 0 .00 0 0 .66 1 0 .9 31
27 . Propylene glycol 0 .794 0 .0 00 0 .6 88 0 .89 9
28 . Glycolate 0 .79 2 0 .00 0 0 .67 8 0 .9 05
29 . -Alan ine 0 .78 6 0 .00 0 0 .67 5 0 .89 83 0 . Sar cosin e 0 .785 0 .000 0 .6 70 0 .9 01
31 . Thr eonate 0 .78 5 0 .00 0 0 .67 1 0 .9 00
32 . Tran s-4 -hydr oxy-l-p rolin e 0 .7 84 0 .0 00 0 .67 1 0 .8 98
33 . Pim ela te 0 .783 0 .000 0 .656 0 .91 1
34 . Cysta th ion ine 0 .7 81 0 .00 0 0 .66 8 0 .89 4
35 . Alan in e 0 .779 0 .0 00 0 .6 64 0 .89 4
36 . Succinylace ton e 0 .777 0 .000 0 .661 0 .89 3
37 . Tr im ethylam ine n -oxide 0 .77 7 0 .000 0 .659 0 .8 95
38 . 2 -Oxocapr oate 0 .776 0 .000 0 .65 9 0 .8 93
39 . Ser in e 0 .7 76 0 .0 00 0 .6 54 0 .8 98
40 . 2 -Methylglu tar a te 0 .77 5 0 .00 0 0 .66 0 0 .8 90
41 . Meth ion ine 0 .768 0 .001 0 .64 3 0 .8 92
42 . Glycylp roline 0 .7 67 0 .0 01 0 .6 48 0 .8 86
43 . 5 -Am in olevu lina te 0 .765 0 .001 0 .650 0 .87 9
44 . 2 -Phosp hoglycer ate 0 .76 3 0 .00 1 0 .64 4 0 .8 81
45 . 2 -Oxobu tyra te 0 .75 6 0 .00 1 0 .63 4 0 .87 8
4 6 . Glutaric acid monomethyl ester 0 .756 0 .001 0 .6 34 0 .8 78
47 . 1 ,3 -Dim e thylu rate 0 .75 5 0 .00 1 0 .63 9 0 .8 71
48 . 3 -Methylglu tar a te 0 .75 4 0 .00 1 0 .62 8 0 .8 80
49 . Fr uctose 0 .7 53 0 .0 01 0 .6 21 0 .8 85
50 . Ascorbate 0 .750 0 .00 1 0 .62 1 0 .8 79
51 . Ethylene glycol 0 .748 0 .0 01 0 .62 3 0 .87 3
52 . Asp ar tate 0 .74 7 0 .00 1 0 .62 2 0 .87 253 . Galacton ate 0 .7 43 0 .0 02 0 .61 5 0 .8 70
54 . Methylsuccin ate 0 .739 0 .002 0 .611 0 .86 6
55 . Xylose 0 .7 36 0 .0 02 0 .6 06 0 .8 67
56 . Prolin e 0 .73 3 0 .00 3 0 .61 2 0 .8 55
57 . Isocitrate 0 .732 0 .00 3 0 .60 4 0 .8 60
58 . N-Car bam oylasp ar tate 0 .731 0 .003 0 .590 0 .87 3
59 . 2 -Ethylacr yla te 0 .729 0 .003 0 .59 3 0 .8 65
60 . 2 -Oxoisocapr oate 0 .724 0 .00 4 0 .58 3 0 .865
61 . Sucrose 0 .72 4 0 .0 04 0 .5 73 0 .8 75
62 . Glucose 0 .722 0 .00 4 0 .59 5 0 .8 49
63 . 4 -Hydroxybu tyrate 0 .720 0 .005 0 .5 89 0 .85 164 . N-Car bam oyl- -a lan in e 0 .720 0 .004 0 .578 0 .86 2
65 . Valine 0 .72 0 0 .00 5 0 .58 0 0 .8 59
66 . Galact itol 0 .71 9 0 .00 5 0 .58 9 0 .8 48
67 . Man n itol 0 .710 0 .007 0 .5 67 0 .8 53
68 . Tr im e thylam in e 0 .710 0 .00 7 0 .57 1 0 .8 50
69 . Isobu tyrate 0 .70 4 0 .00 8 0 .57 3 0 .8 36
70 . N,N-d im e thylglycin e 0 .70 3 0 .00 9 0 .55 9 0 .84 7
71 . Methylguan id ine 0 .70 1 0 .00 9 0 .561 0 .84 2
7 2 . Hom oser in e 0 .7 00 0 .0 10 0 .5 57 0 .8 43
Table: 4, show s m etabolite at th eir 95 % con fidence inte rval, the ir are a un der t he ROC curve w hich is above
0.70 along w ith the p values.
7/30/2019 Metabo No Mics
12/13
Table 5, Final Metabolome with highest degrees of accuracy
Table 5. Lists the final me tabolom e w ith highest degrees of accura cy to discrimina te h ep a-
tocellular carcinom a from th ose pa tients with chron ic l iver d isease or ap pare ntly norm al
liver
Table 6, Metabolites a nd their mea sure s of accuracy on pa rallel combinations
IJPCR VOL. 1 (1 ) JULY 20 12 w w w.ijp cr.n e t 18
Table 6 , show s th e p arallel comb ination of metabolites in table 4 . On p arallel combina tions s om e of
the me tabolites sh ow im pro ved sen sitivity and spe cificity.
Sl n o Me tabolites Cu t offs Sen sit ivity Sp ecificity LR LR
(+) ( -)
1 . 1 ,6 -An hydr o- -D-glucose 4 .5 6 52 8 0 .0 0 8 5 .42 5 .4 9 0 .2 3
2 . Galacta r ate 4 .8 22 3 90 .00 79 .17 4 .3 2 0 .1 33 . 1 ,3 -Dihydr oxyaceton e 1 .4 6 28 7 0 .0 0 8 3 .33 4 .2 0 0 .3 6
4 . 2 - Bu tan on e 5 .5 89 8 8 5 .0 0 7 7 .0 8 3 .7 1 0 .1 9
5 . 2 -Hydr oxyglu tarate 24 .71 7 9 0 .0 0 7 5 .0 0 3 .6 0 0 .1 3
6 . 2 -Oxoisocap r oate 3 .50 5 6 0 .0 0 8 3 .3 3 3 .6 0 0 .4 8
7 . Citr a te 6 .9 83 5 8 5 .0 0 7 5 .0 0 3 .4 0 0 .2 0
8 . Cystath ion in e 8 .0 30 4 70 .00 79 .1 7 3 .3 6 0 .3 8
9 . Acetoace ta te 3 .0 80 8 95 .00 70 .83 3 .2 6 0 .0 7
10 . Succin a te 1 .7 02 7 80 .00 75 .00 3 .2 0 0 .2 7
11 . Tar tar ate 5 .83 7 6 85 .00 72 .92 3 .1 4 0 .2 1
1 2 . Tr im ethylam in e N-oxide 5 .8 37 6 8 5 .0 0 7 2 .9 2 3 .1 4 0 .2 1
13 . Pyr uvate 2 .9 39 75 .00 75 .00 3 .0 0 0 .3 3
Sl no Metabolite Combinat ions Sensit ivity Specificity LR+ LR-
1 . Ser in e an d 2 -Ethylacr ylate 100 68 .75 3 .20 0
2 . Glycylp roline an d Methylam in e 100 64 .58 2 .82 0
3 . Methylam in e and Galactarate 100 64 .5 8 2 .82 04 . Methylam ine and Prop ion ate 1 00 5 0 2 .0 0
5 . 2 -Oxocap roate and Methylam ine 100 50 2 .0 0
6 . 2 -Hydr oxyglu tarate and Galactarate 9 5 70 .83 3 .2 6 0 .07
7 . Guan idoacetate and 1,6-Anhydro --D-glucose 9 5 68 .75 3 .05 0 .07
8 . 2-Methylglutarate and 1,6-Anhydro--D-glucose 9 5 66 .67 2 .85 0 .08
9 . Ace ton e an d cysta th ion ine 9 5 62 .5 0 2 .53 0 .0 6
10 . 1 ,6 -An hydr o- -D-glucose an d Ethylen e glycol 9 5 60 .42 2 .4 0 0 .08
11 . Ace tone an d 2 ethylacr ylate 9 5 50 1 .90 0 .10
12 . Galactorate and Cysta th ion in e 90 75 3 .6 0 0 .13
13 . 2 hydroxyglu tar ate and p im e late 9 0 66 .6 7 2 .7 0 0 .15
14 . Galactarate and Tr im e thylam in e N-oxide 90 6 4 .58 2 .5 4 0 .1115 . Succin ate an d Cystath ion in e 85 75 3 .40 0 .2 0
16 . Bu tan one an d pyr uvate 85 70 .83 2 .9 1 0 .21
17 . 3 hydr oxy-3 -m e thylglu tar ate an d Succin ate 8 5 70 .83 2 .9 1 0 .21
18 . Pim e late and Cysta th ion in e 8 5 66 .6 7 2 .55 0 .2 3
19 . Alan in e an d Pim elate 85 64 .58 2 .40 0 .23
20 . 2 -m e thylglu tarate and guan idoacetate 8 5 60 .42 2 .1 5 0 .17
21 . Pyr uvate an d Cysta th ion ine 80 75 3 .2 0 0 .27
22 . Cysta th ion ine an d Succinylace ton e 75 7 0 .83 2 .57 0 .3 5
7/30/2019 Metabo No Mics
13/13
Figure 1, Exper imen tal spectru m after p rofiling
Figure 1; shows an experimental spectrum after profiling. The green line corresponds to the ex-
perimen tal spectral line of serum, under wh ich som e of the reference comp ounds are m atched in
correspon ding colors .
Figur e 2 , Sam ple ROC cur ve p lotte d for 1 ,6Anhydr o -D-Glucose
Figure 2, show s a s am ple ROC cur ve plotted for 1,6Anhydro -D-Glucose an d its op tim al
cut off along w ith its m eas ur es of diagno stic accuracy.
,
IJPCR VOL. 1 (1 ) JULY 20 12 w w w.ijp cr.n e t 19