Serum Concentrations of Insulin-like Growth Factor-I and Insulin-likeGrowth Factor Binding Protein-3 in Patients with Chronic Hepatitis

A. Okan, A. Comlekci, H. Akpınar, I˙. Okan, S. Yesil, E. Tankurt & I ˙. SimsekDept. of Gastroenterology & Dept. of Endocrinology, Dokuz Eylu¨l University Medical School, I˙zmir,Turkey

Okan A, Comlekci A, Akpınar H, Okan I˙, Yesil S, Tankurt E, S¸ imsek I. Serum concentrations of insulin-like growth factor-I and insulin-like growth factor binding protein-3 in patients with chronic hepatitis.Scand J Gastroenterol 2000;35:1212–1215.

Background: Insulin-like growth factor-I is a liver-derived humoral factor, which has important anabolicand metabolic actions and is predominantly bound by insulin-like growth factor binding protein-3. Lowserum concentrations of both insulin-like growth factor-I and insulin-like growth factor binding protein-3have been reported in patients with chronic liver disease, especially cirrhosis, but their conditions inchronic hepatitis are uncertain. The aim of this study was to evaluate the effect of chronic hepatitis onserum concentrations of insulin-like growth factor-I and insulin-like growth factor binding protein-3 andtheir association with hepatic inflammation activity and fibrosis.Methods: Serum insulin-like growthfactor-I and insulin-like growth factor binding protein-3 were measured by RIA (ng/ml) in 17 patients withmild to severe chronic viral hepatitis (12 chronic hepatitis C, 5 chronic hepatitis B) and 16 healthy subjects.The hepatic inflammation activity and the severity of fibrosis were evaluated using Desmet classification.Results:Both insulin-like growth factor-I and insulin-like growth factor binding protein-3 levels did notcorrelate with inflammation activity, fibrosis or transaminase levels. In the chronic hepatitis group, insulin-like growth factor-I levels were significantly higher than the control group (mean, 263.8� 27.33 versus127.14� 10.83 ng/ml,P< 0.001, respectively), whereas insulin-like growth factor binding protein-3levels were significantly lower when compared with the controls (1643.47� 60.68 versus 2728.87� 284.61 ng/ml,P< 0.05, respectively).Conclusions:These results suggest that the concomitant states ofserum insulin-like growth factor-I and insulin-like growth factor binding protein-3 levels in patients withchronic hepatitis may be different from cirrhotic patients and high serum IGF-I levels may be a specificfinding of the stage of chronic hepatitis before developing cirrhosis.

Key words: Chronic hepatitis; insulin-like growth factor-I; insulin-like growth factor binding protein-3

Abdullah Okan, M.D., Dept. of Gastroenterology, Dokuz Eylu¨ l University Hospital, 35340 I˙nciraltı,Izmir, Turkey (fax:�90 232 259 9723)

Insulin-like growth factor-I (IGF-I) is a liver-derivedhumoral factor, which has important anabolic andmetabolic actions and is predominantly bound by

insulin-like growth factor binding protein-3 (IGFBP-3) andthey are mainly synthesized in the liver (1, 2). Low serumconcentrations of both IGF-I and IGFBP-3 have been reportedin patients with chronic liver disease, especially cirrhosis, buttheir conditions in chronic hepatitis are uncertain. The aim ofthis study was to evaluate the effect of chronic hepatitis onserum concentrations of IGF-I and IGFBP-3 and theirassociation with hepatic inflammation activity and fibrosis.

Materials and Methods

PatientsSeventeen patients with mild to severe chronic viral

hepatitis (12 chronic hepatitis C, 5 chronic hepatitis B; 11

female, 6 male; mean age, 50.8� 13.8 years) and 16 healthysubjects as a control group (8 female, 8 male; mean age,54.4� 10.3 years) were included in the study. All patientshad the histological diagnosis of chronic hepatitis and theiralanine aminotransferase levels (ALT) were at least above�1.5 times the normal upper limit at the time of obtainingserum samples for the detection of IGF-I and IGFBP-3. Inchronic hepatitis C and B patients, serum HCV–RNA wasdetected by polymerase chain reaction and HBV–DNAdetected by chemiluminescent molecular hybridization.

The hepatic inflammation activity and the severity offibrosis were evaluated using Desmet classification (3). Sincediabetes mellitus has an effect on serum IGF-I levels (4),diabetic patients were excluded from the study and allparticipants had no diabetes mellitus or impaired glucosetolerance, as assessed by the new criteria for diagnosis ofdiabetes mellitus reported from the Expert Committee (5) and

ORIGINAL ARTICLE

2000 Taylor & Francis

Scan

d J

Gas

troe

nter

ol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 10

/30/

14Fo

r pe

rson

al u

se o

nly.

none of the patientswith chronic hepatitis had any otherunderlyingdiseaseandwereunderspecifictherapyfor theirhepatitisat entry to the study,suchas interferontreatment.Only 5 chronichepatitisC and3 chronichepatitisB patientsreceivedinterferontherapyuntil at least9 monthsago.AllcontrolsubjectshadnormalALT levelsandwerenegativeforHBsAg and anti-HCV, as assessedby second and thirdgenerationELISA tests,respectively.

Measurementsof IGF-I and IGFBP-3 in serumIGF-I wasdeterminedby immunoradiometricassayusing

mouse monoclonal antibodies (Immunotech, Marseilles,France).In order to releaseIGF-I from its binding proteins,serumsamples,extractedwith acid ethanol,and standardswere incubatedin tubes coatedwith the first monoclonalantibodyin thepresenceof a second125I-labeledmonoclonalantibodyasa tracer.After incubation,the liquid contentsofthe tubes were removed and bound radioactivity wasmeasured.The detection limit was 12ng/ml. Intra-assaycoefficients of variation were 8.9%, 5.7% and 7.1% atanalytic levels of 95.4, 360.9and 1212ng/ml, respectively,and inter-assayvariationswere 11.9%,8.6% and 16.7% at87.2,361.2and1211ng/ml, respectively.

IGFBP-3wasmeasuredby immunoradiometricassayusingmousemonoclonalantibodiesdirectedagainsttwo differentepitopesof the molecule(Immunotech,Marseilles,France).Samplesandstandardswereincubatedin tubes,coatedwiththe first monoclonal antibody, with a second125I-labeledmonoclonalantibody.The liquid contentsof the tubeswereaspiredafter incubationand bound radioactivity was mea-sured.The detectionlimit of the assaywas 50ng/ml. Theintra-assaycoefficientsof variation were 10.2%, 5.6% and4.6% at levels of 756, 1739 and 3780ng/ml, respectively.Inter-assaycoefficientsof variation were 12.4%,8.2% and6.3%at 756,1739and3780ng/ml, respectively.

StatisticalanalysisStatistical analysiswas done with a chi-squaretest for

qualitativevariablesandwith the Mann–WhitneyU test forquantitativevariables.CorrelationsamongIGF-I, IGFBP-3,ALT, histological grading and staging in chronic hepatitispatientswereevaluatedby SpearmanandPearsoncoefficientsof correlation.Statisticalsignificancewassetat P< 0.05.

Results

Characteristicsof thestudygroupsaregivenin TableI. Bothstudygroupsweresimilar with regardto age,sexandbodymassindex.Both IGF-I andIGFBP-3levelsdid not correlatewith inflammation activity, fibrosis or transaminaselevels(P> 0.05). In the chronic hepatitis group, IGF-I levelswere significantly higher than the control group (mean,263.8� 27.33 versus127.14� 10.83ng/ml, P< 0.001, re-spectively),whereasIGFBP-3levelsweresignificantlylowerwhen comparedwith the controls (1643.47� 60.68 versus2728.87� 284.61ng/ml,P< 0.05,respectively)(Figs.1 and2).

TableI. Characteristicsof thestudygroups

Chronichepatitis Healthysubjects P

Sex(male/female) 6/11 8/8 NS‡Age (years),mean� s* 50.8(25–74)� 13.8 54.4(35–70)� 10.3 NSBody massindex (kg/m2), mean� s 27.2� 3.3 26.7� 4.8 NSALT† (IU/l), mean� s 73.7� 19 19.4� 6.2 <0.00001Histologicalgradingof hepatitis

Mild 6 — —Moderate 9 — —Severe 2 — —

Histologicalstagingof hepatitisNo fibrosis 1 — —Mild fibrosis 8 — —Moderatefibrosis 6 — —Severefibrosis 2 — —

* s= standarddeviation.† ALT = alanineaminotransferase.‡ NS= not significant.

Fig. 1. Box-plot graphicsof serumIGF-I (ng/ml) levels in patientswith chronichepatitisandhealthysubjects(control).The lower andupper edges of the boxes show 25th and 75th percentiles,respectively; vertical lines extending from boxes show 1st and99th percentiles;horizontallines in boxesindicatemedianvalues.

ScandJ Gastroenterol2000(11)

IGF-I and IGFBP-3 in chronic hepatitis 1213

Scan

d J

Gas

troe

nter

ol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 10

/30/

14Fo

r pe

rson

al u

se o

nly.

Discussion

Many studies have shown that both IGF-I and IGFBP-3concentrationsare low in patientswith liver cirrhosis(6–9)andlow IGF-I levelscorrelatewith thedegreeof liver failure(10). Therefore,IGF-I and IGFBP-3 have beenreasonablyproposedas markersof hepatocellularcapacity(10–12).Inthis context, Caregaroet al. (13) demonstratedthat IGF-Isignificantlyincreasedafterorthotopicliver transplantationincirrhoticpatientswhencomparedto its baselinevalues.In therelatedtrialsconductedupuntil thepresent,thestateof IGF-Iand IGFBP-3 have not been widely examinedin chronichepatitis,especiallyin chronicviral hepatitis,exceptfor onlya few studies.Khoslaet al. (14) foundno alterationin serumIGF-I levels of patientswith hepatitis C-associatedosteo-sclerosiswhencomparedto controlgroups,includingnormalhealthy individuals, hepatitis B patients and hepatitis Cpatients without osteosclerosis,whereas serum IGFBP-3levels were significantly lower in both hepatitisC patientswith or without osteosclerosisandhepatitisB patientsthaninhealthy ones; there were also no significant differencesinIGF-I levelsamongcontrol groups.But, in this study,therewas no laboratorydata regardinghistological examinationsincluding gradingandstagingandALT levels in hepatitisBandC patients,exceptfor serologicalevidenceof hepatitisBandC viruses.Also, noneof thesepatientswassymptomatic.In another study, Del Monte et al. (15) demonstratedasignificant increasein serumIGF-1 levels during interferontreatmentin patientswith chronicactivehepatitisandit wasinverselycorrelatedwith ALT levels.However,the authorshad not comparedthe pre-treatmentlevels of IGF-I withnormal subjectsso it was not certain whetherthe baselinelevels of IGF-I in chronic hepatitisC significantly differedfrom normalones.

In contrastto the reportedlow IGF-I levels in cirrhoticpatients,we found high serumIGF-I levels in patientswith

chronichepatitiswhencomparedto thehealthycontrols,butIGFBP-3levelswerelower thanthecontrols,similar to thosewith cirrhosis.In ourchronichepatitispatients,theabsenceofaninversecorrelationbetweenfibrosisscoresandIGF-I levelssuggeststhat the reasonfor low concentrationsof IGF-I incirrhoticsmay not be directly associatedwith the degreeoffibrosisor IGF-I levelswill notbeginto decreaseunlessthereis sufficientfibrosisto inducehepatocellulardysfunction.

Theseresultssuggestthat theconcomitantstatesof serumIGF-I and IGFBP-3 levels in patientswith chronichepatitismaybedifferentfrom cirrhotic patientsandhigh IGF-I levelsmay be a specific finding of the stageof chronic hepatitisbefore developingcirrhosis, but its pathophysiologicalim-portanceremainsunclear.Recentstudieshave shown thatIGF-I is mainly synthesizedby parenchymalliver cells andIGFBP-3mainly by Kupffer cells (2,16,17). Furtherstudiesareneededto addresswhethertheincreasedIGF-I mightbeanindicatorof hepatocellulardamageor liver regenerationandwhetherthe decreasedIGFBP-3 might reflect that Kupffercells are also affectedby hepaticinflammationlike hepato-cytes,andits synthesisis reduced,or whetherit is bindingtoexcessIGF-I in thecirculation.SinceIGF-I alsohasautocrineandparacrineeffectsotherthanits endocrineeffect (18), thedeterminationof IGF-I andIGFBP-3levelsin tissuewith theirconcomitantserumconcentrationswill provide more infor-mationabouttheir statusin chronichepatitis.

References

1. DaughadayWH, RotweinP. Insulin-like growthfactorsI andII.Peptide,messenger ribonucleicacid andgenestructures,serum,andtissueconcentrations.EndocrRev 1989;10:68–91.

2. Arany E, Afford S, StrainAJ, WinwoodPJ,Arthur MJ, Hill DJ.Differential cellular synthesis of insulin-like growth factorbindingprotein-1(IGFBP-1)andIGFBP-3within humanliver. JClin EndocrinolMetab1994;79:1871–6.

3. DesmetVJ, GerberM, HoofnagleJH, MannsM, ScheuerPJ.Classification of chronic hepatitis: diagnosis, grading andstaging.Hepatology1994;19:1513–20.

4. Tan K, BaxterRC. Seruminsulin-like growth factor I levels inadult diabetic patients: the effect of age. J Clin EndocrinolMetab1986;63:651–5.

5. Reportof the expertcommitteeon the diagnosisandclassifica-tion of diabetesmellitus DiabetesCare 1998;21Suppl 1:S5–S19.

6. SheppardMS, Minuk GY, BhaumickB, Bala RM. Insulin-likegrowth factors (IGF) in liver disease:differential changesofIGF-I andIGF-II. Clin InvestMed 1987;10:49–53.

7. DonaghyA, RossR, GimsonA, HughesSC,Holly J, WilliamsR. Growth hormone,insulin-like growth factor-1, and insulin-like growth factor binding proteins 1 and 3 in chronic liverdisease.Hepatology1995;21:680–8.

8. Moller S, Juul A, Becker U, Flyvbjerg A, SkakkebaekNE,HenriksenJH. Concentrations,release,and disposalof insulin-like growth factor (IGF)-binding proteins(IGFBP), IGF-I, andgrowth hormone in different vascular beds in patients withcirrhosis.J Clin EndocrinolMetab1995;80:1148–57.

9. Assy N, HochbergZ, Amit T, Shen-OrrZ, Enat R, BaruchY.Growth hormone-stimulatedinsulin-like growth factor (IGF) IandIGF-bindingprotein-3in liver cirrhosis.J Hepatol1997;27:796–802.

10. CaregaroL, Alberino F, Amodio P,Merkel C, Angeli P,PlebaniM, et al. Nutritional andprognosticsignificanceof insulin-like

Fig. 2. Box-plot graphics of serum IGFBP-3 (ng/ml) levels inpatientswith chronic hepatitisand healthysubjects(control). Thelower andupperedgesof theboxesshow25thand75thpercentiles,respectively;vertical lines extendingfrom boxesshow1st and99thpercentiles;horizontallines in boxesindicatemedianvalues.

ScandJ Gastroenterol2000(11)

1214 A. Okanet al.

Scan

d J

Gas

troe

nter

ol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 10

/30/

14Fo

r pe

rson

al u

se o

nly.

growth factor 1 in patients with liver cirrhosis. Nutrition1997;13:185–90.

11. Moller S, Becker U, Juul A, SkakkebaekNE, ChristensenE.Prognosticvalueof insulin-like growth factor I and its bindingprotein in patientswith alcohol-inducedliver disease.EMALDgroup.Hepatology1996;23:1073–8.

12. ShaarawyM, Fikry MA, MassoudBA, Lotfy S. Insulin-likegrowth factor binding protein-3: a novel biomarker for theassessmentof the synthetic capacity of hepatocytesin livercirrhosis.J Clin EndocrinolMetab1998;83:3316–9.

13. CaregaroL, Alberino F, Angeli P, GattaA. Insulin-like growthfactor 1 (IGF-1) in liver cirrhosis: a marker of hepatocellulardysfunction?J Hepatol1998;29:342.

14. KhoslaS,HassounAA, BakerBK, Liu F, Zein NN, WhyteMP,et al. Insulin-like growthfactorsystemabnormalitiesin hepatitis

C-associatedosteosclerosis. Potential insights into increasingbonemassin adults.J Clin Invest1998;15:2165–73.

15. Del MonteP,BernasconiD, De ConcaV, RandazzoM, MeozziM, BadaraccoB, et al. Endocrineevaluationin patientstreatedwith interferon-alphafor chronic hepatitisC. Horm Res1995;44:105–9.

16. Chin E, Zhou J, Dai J, Baxter RC, Bondy CA. Cellularlocalizationandregulationof geneexpressionfor componentsofthe insulin-like growth factor ternarybinding protein complex.Endocrinology1994;134:2498–2504.

17. Villafuerte BC, Koop BL, PaoCI, Gu L, BirdsongGG, PhillipsLS. Cocultureof primary rat hepatocytesandnon-parenchymalcells permits expressionof insulin-like growth factor bindingprotein-3in vitro. Endocrinology1994;134:2044–50.

18. SaraVR, Hall K. Insulin-like growth factorsand their bindingproteins.PhysiolRev 1990;70:591–614.

Received16 November1999Accepted14 April 2000

ScandJ Gastroenterol2000(11)

IGF-I and IGFBP-3 in chronic hepatitis 1215

Scan

d J

Gas

troe

nter

ol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 10

/30/

14Fo

r pe

rson

al u

se o

nly.