AISF position paper on nonalcoholic fatty liver disease (NAFLD): Updates and future directions

(1)

ContentslistsavailableatScienceDirect

Digestive

and

Liver

Disease

j o ur n a l ho me p a g e :w w w . e l s e v i e r . c o m / l o c a t e / d l d

Position

Paper

AISF

position

paper

on

nonalcoholic

fatty

liver

disease

(NAFLD):

Updates

and

future

directions

夽

The

Italian

Association

for

the

Study

of

the

Liver

(AISF)

a

r

t

i

c

l

e

i

n

f

o

Articlehistory:

Received29December2016

Receivedinrevisedform8January2017 Accepted10January2017

Availableonline23January2017 Keywords: Diagnosis Epidemiology Genetics Management Naturalhistory Physiopathology

a

b

s

t

r

a

c

t

Thisreviewsummarizesourcurrentunderstandingofnonalcoholicfattyliverdisease(NAFLD),a multi-factorialsystemicdiseaseresultingfromacomplexinteractionbetweenaspeciﬁcgeneticbackground andmultipleenvironmental/metabolic“hits”.

Theroleofgutmicrobiota,lipotoxicity,inﬂammationandtheirmolecularpathwaysisreviewed in-depth.WealsodiscusstheepidemiologyandnaturalhistoryofNAFLDbypinpointingtheremarkably highprevalenceofNAFLDworldwideanditsinherentsystemiccomplications:hepatic(steatohepatitis, advancedﬁbrosisandcirrhosis),cardio-metabolic(cardiovasculardisease,cardiomyopathy,arrhythmias andtype2diabetes)andneoplastic(primarylivercancersandextra-hepaticcancers).

Moreover,wecriticallyreportonthediagnosticroleofnon-invasivebiomarkers,imagingtechniques andliverbiopsy,whichremainsthereferencestandardfordiagnosingthedisease,butcannotbeproposed toallpatientswithsuspectedNAFLD.

Finally,themanagementofNAFLDisalsoreviewed,byhighlightingthelifestylechangesandthe pharmacologicaloptions,withafocusontheinnovativedrugs.

Weconcludethattheresultsofongoingstudiesareeagerlyexpectedtoleadtointroduceintothe clinicalarenanewdiagnosticandprognosticbiomarkers,preventionandsurveillancestrategiesaswell astonewdrugsforatailoredapproachtothemanagementofNAFLDintheindividualpatient.

1. Introduction

Nonalcoholicfattyliverdisease(NAFLD)hasprompteda grow-ingclinicalandresearchinterestoverthelast25years.Paramount information hasprogressivelyaccumulatedonthegenetics and behavioralriskfactorsfordiseasedevelopmentandprogression [1,2],onhepaticandextra-hepaticcomplications[3,4]andon puta-tivetreatmentstrategies[5,6].Nevertheless,theburdenofdisease isstillincreasing,mainlyduetotherisingtideofobesityandtype 2diabetesmellitus(T2DM)epidemics(“diabesity”)andthelackof effectivetreatmentoptions.NAFLDwhich,inaclosedloopfuelsthe

TheItalianAssociationfortheStudyoftheLiver(AISF)AmedeoLonardoa,1_,_Fabio

Nascimbenib,1_, _Giovanni _Targherc_,_Mauro _Bernardid_, _Ferruccio_Boninoe_,

Elisa-bettaBugianesif_,_Alessandro_Casinig_,_Amalia_Gastaldellih_,_Giulio_Marchesinii_,_Fabio

Marraj_,_Luca_Mielek_,_Filomena_Moriscol_,_Salvatore_Pettam_,_Fabio_Piscaglian_,

Gian-lucaSvegliati-Baronio_,_Luca_Valentip_,_Stefano_Bellentaniq,∗ a_Division_of_Internal Medicine,NOCSAE,AziendaOspedaliero-Universitaria,Modena,Italy

b-q_See_Appendix_A.

夽 Reportof2015AISFMonothematicConferenceonNAFLD.

∗ Correspondingauthorat:CapoServiziodiGastroenterologiaedEpatologia, Clin-icaSantaChiara,ViaFranscini4,6610Locarno,Switzerland.

E-mailaddress:[email protected].

1 _These_authors_contributed_equally_to_this_article.

“diabesity”epidemics[7],maystartveryearly–eveninutero[8,9] –andthelongertheexposure,thehighertherisktodevelopan advanceddiseaseanditscomplications[10].

So,whatcanbedonetohaltthedevelopmentorreducethis bur-denofdisease?Europeanclinicalguidelinesonthemanagementof NAFLDhavebeenrecentlypublishedbythethreesistersocietiesof LiverDisease,DiabetesandObesity[11].Atthetimeofpublication ofsuchguidelines,theparticipatingExpertsraisedconsiderable concernabout thedifﬁcultiesexpected in conducting universal screeningandappropriatesurveillancestrategiesandfollow-upin suchapotentiallyhugepopulationofindividuals.Indeed,the num-berofindividualsatriskseemstoolargetobeaffordablebythe Nationalhealthcaresystems,butselectioncriteriadonotguarantee satisfactorysensitivityandspeciﬁcitytoidentifydisease progres-sion[12].Surrogate,non-invasivemarkersofNAFLDcanbeused fortheirnegativepredictivevaluetosparepatientsfrom under-goingliverbiopsy;however,theirresultsmayassistphysiciansin dictatingtheprognosis,notinguidingtreatmentstrategies,given thatvalidatedpharmacologicalinterventionsarelacking[13].

Disease control may,therefore, only be achieved by appro-priate societal interventions aimed at keeping at bay the behavioralriskfactorsalsoinvolvedinthepathogenesisof “dia-besity”. Nonalcoholic steatohepatitis-cirrhosis (NASH-cirrhosis) http://dx.doi.org/10.1016/j.dld.2017.01.147

(2)

andNASH-hepatocellularcarcinoma(NASH-HCC)areonlythe“tip oftheiceberg”oftheunhealthyconsequencesofnon-communicable diseases[14].

Againstthisbackground,this updatedpositionpaperaimsto summarizingthechieftopicsdiscussed duringtheAssociazione ItalianaperloStudiodelFegato(AISF)SingleTopicConferenceon NAFLDheldinModena(October8–10,2015)inmemoryoflate Pro-fessorPaolaLoria,whowastheCoordinatoroftheﬁrstAISFclinical practiceguidelinesformanagementofNAFLD[15].

2. PathophysiologyofNAFLD 2.1. Pathogenesis

NAFLDis a multi-factorial disease resultingfrom a complex interaction of environmental “hits” and a genetic background (Fig.1).Ahigh-caloriediet,oftencoupledwithasedentary behav-ior,contributetothedevelopmentofNAFLD,bothdirectlyandvia weightgain.Dietaryexcessofsaturatedfatsandrefined carbohy-drateshasbeenassociatedwithNAFLDandahighfructoseintake mayincreasetheriskofNASH[16,17].Developmentand progres-sionofNAFLDarestronglyassociatedwithinsulinresistance(IR) andmetabolicsyndrome(MetS)components,particularly abdom-inalobesityandT2DM[18].ThemostcommoncauseofNAFLD is analtered whole-bodyenergetic homeostasis,due tocaloric intakeexceedingcaloricexpenditure,withconsequentspillover ofextra-energyin theformof non-esterifiedfattyacids(NEFA) fromvisceraladiposetissue intoectopicfatdepots,suchasthe liver,skeletalmusclesand pancreas[19].NAFLDwill invariably developwhentherateofhepatictriglycerideflowingtotheliver viathebloodstreamorsynthesizedwithintheliverexceedsthe rateofhepatictriglycerideoxidationandVLDLsecretionintothe bloodstream[19].Approximately60%ofhepaticlipidsderivesfrom increasedperipherallipolysisoftriglycerides(duetoadipose tis-sueIRandfailuretoadequatelysuppressperipheraltriglyceride lipolysis),whiledietaryfatsandsugarscontributeapproximately 35–40%[20].Theliveritselfmayalsocontributetosteatogenesisby synthesizingtriglyceridesfromdietarycarbohydratesthroughde novolipogenesis.Thecontributionofdenovolipogenesistoliver fatcontentislessthan5%inhealthysubjectsandmayincreaseto approximately25%inNAFLDpatients[20].Intra-hepatocytic accu-mulationofdiacylglycerolintermediatesimpairshepaticinsulin signalingandfuelsgluconeogenesis,sopromotinghyperglycemia and predisposing to the development of T2DM [21]. Increased amounts of circulating and intracellular NEFAs are also associ-atedwithanincreaseinnuclearfactorkappa-B(NF-␬B),eventually leadingtotheexpandedanddysfunctionaladiposetissue overpro-ducingmultiplepro-inflammatorycytokinesandunder-producing anti-inflammatoryadipokines,suchasadiponectinwhich, collec-tively,mayfurtherdictateNAFLDprogression[22,23].

2.2. Lipotoxicityandinﬂammation

AsshowninFig.1,NASHprogressionresultsfromnumerous eventsoriginatingwithintheliverandindistalorgans,including thevisceraladiposetissueandthegastrointestinaltract[24]. Fat-induceddamagetothehepatocytes(lipotoxicity),ismorelinked totheabundanceofspeciﬁctoxiccompounds,suchasNEFAsand ceramides,thantothetotalamountofstoredfat[25].Evidence fromgeneticstudies,ontheotherhand,suggeststhattheamount offataccumulationisimportant[26].Toxiclipidscandetermine cellinjurythroughavarietyofmechanisms,includingincreased oxidative stress and mitochondrial dysfunction. Saturated fatty acidsareincreasedinNASH[24,25]andinduceinﬂammationand hepatocyteapoptosisby activatingJunN-terminal kinase (JNK)

andmitochondrialpathways[27].Freecholesterolisaprominent mechanismforNASHdevelopmentandprogression.Interestingly, necroptosishasbeenrecentlydescribedasacelldeathmechanism potentiallyinvolvedinlipotoxicity,whichismorphologically com-parabletonecrosis,thoughcharacterizedbydeﬁnitebiochemical pathways[28].

Endoplasmic reticulum (ER) stress also takes part in NASH pathogenesis,astheresultoftheinductionoftheunfoldedprotein response, which is an adaptive mechanism potentially trigger-ing apoptosis. JNK,an activator of inflammation and apoptosis implicated in NAFLD progression, is one of the major media-tors of ERstress [29]. Hypoxiaperturbs lipid homeostasis and insulinsignalingpathways.Moreover,reducedoxygen availabil-ityinduces secretionofpro-inflammatory cytokines[30].These adverseeffectsaremediatedbytwohypoxia-inducible transcrip-tionfactors(HIF-1␣andHIF-2␣),whichregulatecellularresponse tooxygendeficiency andcanbeactivatedbyadditionalstimuli involvedinNASH,includingoxidativestressorinflammatory sig-nals[30].

Chronic inflammationis a key factor in NASH pathogenesis. Kupffercellactivationoccursatanearlystage,andprecedesthe recruitmentofothercells.Attentionhasbeenpaidtothedifferent phenotypesofKupffercells,i.e.,M1andM2,considered primar-ilyimmuno-regulatory[31].Hepatocytedeathisastrongtrigger of inflammation and fibrosis, through signaling pathways that include tumor necrosis factor (TNF)-related apoptosis-inducing ligandreceptor,FasandTNFreceptor,andpromotetheexpression ofseveralcytokinesandchemokines.Differentiationtowarda pro-inflammatory ‘M1phenotype’ isdriven by pathogen-associated molecular patterns (PAMPs) and damage-associated molecular patterns(DAMPs) interactingwithtoll-like receptors(TLR),and inducesexpressionofpro-inflammatoryfactors,suchasinterleukin (IL)-1␤,IL-12,TNF-␣,andchemokinesCCL2andCCL5[32]. Remark-ably, chemokines such as CCL2 and CCL5 may induce hepatic stellatecellactivation,triggeringfibrogenesis.Besidesresidentand infiltratingmacrophages,theroleofotherinflammatorycells,such asneutrophils,lymphocytes,NKcellsanddendriticcellsisactively beingevaluated[24].

TLRsrecognizeendogenousdangersignals,suchasDAMPsor PAMPs[32].TLR-inducedpathwaysplayacentralroleinthe acti-vationofhepaticcells,primarilyKupffercells,butalsohepatocytes andstellatecells.TLR2interactswithmultiplePAMPs,whichare increasedinNAFLD,anditsinhibitionpreventshepatic/systemic IRinhigh-fatdiet–fedmice[32].TLR9isactivatedby unmethy-latedDNA,typicallyexpressedinvirusesandbacteriabutrarein mammaliancells.TLR9downstreamsignalinginvolvesIL-1,and isassociatedwithNASHseverityandfibrosis.Thepivotalroleof TLR4inthepathogenesisofNASHhasbeenshowninTLR4-deficient micethatdisplaylowerlevelsofinflammationandfibrosis.TLR4 isprimarilyactivatedbybacteriallipopolysaccharide,triggering expressionofcytokinesandchemokines(e.g.,TNF-␣,IL-1␤,IL-6 andIL-12)[33].Reactiveoxygenspeciesarealsoinducedin TLR4-activatedKupffercells.TLR4isexpressedbyotherhepaticcells, includingstellatecellsandhepatocytes,whereTLR-4exertsactions relevantforthepathogenesisand progressionofNAFLDtoward fibrosis[33].TLR4-mediatedinflammatoryresponsescanalsobe elicitedbyDAMPsreleasedbynecroticcells,suchashighmobility groupbox-1[34].

TheNOD-likereceptors(NLR),whichparticipateinthe assem-bly of inflammasomes (multi-protein complexes required for initiation of inflammatory signals) play a major role in NASH pathogenesis. Activation of the inflammasome is induced by TLRs together with signals linked tocellular damage, e.g., uric acid, reactive oxygen species or adenosine triphosphate, and results in the secretion of mature IL-1 and IL-18. A role for NLRP3 inflammasome in NAFLD development and progression

(3)

Fig.1. PathogenesisofNASH.Unhealthylifestyleandhighfatdietnotonlyresultsinexcessfataccumulation(includingexpandedvisceralfat)butalsoindysbiosis,where FFAscausesteatosisandexertlipotoxiceffects.Moreover,theinﬂamedadiposetissuereleasesadipokines,includingleptinandresistinassociatedwithincreasedhepatic ﬁbrosis,whilethereleaseofadiponectinisreduced.PAMPs=pathogen-associatedmolecularpatterns;TLRs=toll-likereceptors;DNL=denovolipogenesis.

toNASH hasbeenshown bothin humans and inanimal mod-els[35].However, it hasalsobeen demonstrated that thelack ofNLP3promotesgutdysbiosisand chronicinflammation[36]. ActivationofNLRP3inflammasomehasbeenassociatedwith hep-atocytepyroptosis,arecentlydescribed,inflammasome-mediated, celldeathmechanism.

Differentnuclearreceptorshavebeenimplicatedinthe patho-genesisofNASHanddevelopmentoffibrosis.Importantly,these proteinsmaybetargetedbydrugsalreadypresentintheclinical arena.Thereceptorsonwhichmoreinformationhasaccumulated include thenuclearfarnesoid Xreceptor (FXR), theperoxisome proliferator-activatedreceptor(PPAR)-gammaandthePPAR-alpha [37].Agonistsofthesenuclearreceptorsimprovehepatic steato-sis,inflammationandfibrosis,andarebeingtestedinclinicaltrials (suchasdiscussedatSection5inthisreview).

Adiposetissueisrecognizedasanendocrineorganthatsecretes avarietyofadipokines,whichcontrolsystemicmetabolismand energy homeostasis. Among these, leptin and adiponectin are involvedinthepathogenesisofNAFLDanditsprogressiontoNASH, leptin being identified as a pro-fibrogenic adipokine, whereas adiponectindecreasesinflammationandfibrogenesis[38]. 2.3. Microbiota

Ahugegroupofmicroorganisms,collectivelydeﬁnedasthegut microbiota,colonizesthehumanintestine.Ithasbeenestimated thathumangutharborsatleast100trillionmicrobialcells,which are differently represented along the intestine, reaching the greatest concentration in thelarge bowel [39]. Despite a large inter-individualvariation,adultgutmicrobiotaisfairlystableand isdominatedbythephylaFirmicutesandBacteroidetes,followed byActinobacteriaandVerrucomicrobia[40].

Gut microbiota confers the host several physiologic bene-fits, including immune system development, protection from pathogens,andmaintenanceofintestinalandmetabolic homeo-stasis[41].Quantitativeandqualitativechangesofgutmicrobiota composition, also called dysbiosis, have been associated with thedevelopmentofbothintestinalandextra-intestinaldiseases, includingtheMetS.Aroleofdysbiosisinthedevelopmentof obe-sityandNAFLDhasbeendemonstratedinbothanimalandhuman studies.For example,a seminalstudy showedthatthetransfer ofgutmicrobiotafromobesetoleanindividuals inducedinthe recipients,thesamemetabolicalterationsofthedonors[42].Gut microbiota changes were mainlycharacterizedby an increased Firmicutes/Bacteroidetesratio.However,otherinvestigatorsfailed toconfirmthesefindingsandsuggestedthatmetagenomic-based functionalaspectsofgutmicrobiotaarelikelytobemoreimportant thanphylumspecificity[43].

Severalmechanismsareinvolvedinthedevelopmentof obe-sityanditsassociatedmetaboliccomplications.Inparticular,gut microbiotacantrigger(directlyorviathesynthesisofend-products ofbacterialmetabolism,suchasshort-chain fattyacids), differ-ent signaling pathways, which eventually lead to an increased depositionofperipheralfat[42,44].Furthermore,several intesti-nalmicroorganismscanincreasetheefficiencyofcaloricextraction from the food, thus contributing to the development of obe-sity[45].Lastly,diet-induceddysbiosishasbeenassociatedwith increasedgutpermeability in bothmice[46] and humans[47]. Thiscould leadtoincreased translocation ofbacterial products fromtheintestinallumenintotheportalcirculationthereby trigg-ering chronic inflammation. Moreover, within the liver, these circulatingpathogenscanalsoactivatethefamilyofpattern recog-nition receptors(TLRs)andinducepro-inflammatory pathways, which contributetoliver diseasedevelopmentand progression [48–50].

(4)

2.4. Genetics

Geneticfactorsaccountfor aboutof halfof thevariability in intra-hepaticfatcontent,andﬁbrosistendstobeco-inheritedwith steatosis[26].

Recentgenome-wideassociationstudieshavebeguntounveil thespeciﬁccommongeneticdeterminants ofNAFLD.By farthe mostimportantone,duetohighfrequencyandlargesizeeffect, isthep.I148Mloss-of-functionvariantofthePatatin-like phos-pholipasedomaincontaining-3(PNPLA3)gene.Thisgeneencodes foralipaseinvolvedintheremodelingoflipiddropletsin hepato-cytesandthereleaseofretinolfromhepaticstellatecells[51,52]. Inthepresenceofendogenous/exogenousnoxiousstimuli (obe-sity,IR,alcoholandchronicviralhepatitis),carriersofthePNPLA3 I148Mvariantareatincreasedriskofdevelopingsteatohepatitis, cirrhosisandHCC[52].

The loss-of-function variant p.E167K variant of the trans-membrane 6 superfamily member 2 (TM6SF2)is more rare, is carried by about 10% of individuals, and also predisposes to NAFLDand advanced ﬁbrosis/cirrhosis by reducing lipid secre-tionfromhepatocytes, whileprotecting at thesametime from dyslipidemia and cardiovascular disease [53,54]. Other genetic determinantsofNAFLDincludevariants oftheglucokinase reg-ulatoryprotein,regulatinglipogenesisandthemembrane-bound O-acyltransferasedomaincontaining-7,inﬂuencingphospholipid metabolism[55,56].

Multiplemutationsingenesregulatinglipidsecretion,suchas apolipoproteinB,arealsoassociatedwithincreasedriskofsevere NAFLD,cirrhosisandHCC,andareresponsibleforfamilialcasesof thedisease[57].

Allthesenewdiscoveriesareexpectedtocarryoutan improve-ment in ourability togauge therisk in the individualpatient, identifyingnewdiseasemechanismsandtailoredtherapeutic tar-gets.

3. EpidemiologyandnaturalhistoryofNAFLD 3.1. Epidemiology

Arecentsystematicreviewandmeta-analysisestimatedthat 25%oftheworldadultpopulationhasNAFLDasdiagnosedby imag-ing.AlthoughNAFLDwashighlyprevalentinallcontinents,the highestprevalencerateswerereportedfromSouthAmericaand theMiddleEast,whereasthelowestprevalencewasreportedfrom Africa[58].InItaly,theprevalenceofNAFLDrangesapproximately from20%[59]to25%[60–62].

Demographic(age,sexandethnicity)andmetabolicmodiﬁers (MetScomponents)modulatetheriskofdevelopingNAFLD.The prototypicNAFLDpatientisa60-to-85year-oldHispanicman;in contrast,individualsofEuropeanand,inparticular,Africandescent are more spared from NAFLD [63]. Patients with either severe obesity(∼98%)orT2DM(∼70%)aremaximallypronetoNAFLD development; NAFLD is equally prevalent in either sex among patientswithT2DMwhich,therefore,abrogatesthehigher preva-lenceofthemalesextypically observedinnon-diabeticNAFLD individuals[63].TheriskofNAFLDisalsomaximal(∼85%)among patientswithcombineddyslipidemiaandelevatedserum amino-transferases[64].

TheprevalenceofNASHrangesfromasmanyas59.1%among biopsiedNAFLDpatientsto29.9%forNorthAmericaand6.7%for Asia,respectively,amongpatientswhoarenotcandidatesforliver biopsy[58].

ThepooledregionalNAFLDincidence estimatesforAsia and Israelwerereportedtobeapproximately52per1000and28per 1000person-years,respectively[58].InItaly,apioneeringstudy

performedona cohort of hysterectomizedwomen reportedan incidencerateofNAFLDofapproximately2per1000women/year [65].

Arecentmeta-analysisestimatedthatthepooledmean ﬁbro-sisprogressionrateperyearforpatientswithbiopsy-provenNASH was0.09,andapproximately40%ofthesepatientsprogressedto themoreadvancedstagesovertime[58].Theannualincidenceof HCCinNAFLDpatientswas0.44per1000person-years,whereas the corresponding ﬁgure for patients with NASH was 5.3 per 1000person-years[58].Comparedtonon-NAFLDcontrolsubjects, patientswithNAFLDareexposedtoanalmosttwo-foldincreased riskofliver-relatedmortality[58]andtoasubstantiallyincreased riskofdevelopingfatalandnon-fatalCVDevents[66].

Recentstudieshavedescribed thespeciﬁcfeaturesofNAFLD epidemiologyinleanindividuals[67],elderlypeople[68]and chil-dren[69].

3.2. NaturalhistoryofNAFLD:hepaticrisk

NAFLDcoursehasfourclinical-pathologicalentities:steatosis, NASH,advancedﬁbrosis/cirrhosis,andHCC.Researchhasfocused onriskfactorsforeachofthesestepsandfactorsregulatingthe migrationofpatientsfromeachentitytotheothers[70].

3.2.1. Steatosis

Normalliverisalmosttotallydevoidofanyfatcontent[71]. Con-versely,mostpatientswithNAFLDwillhavesimplesteatosis,(fatty changesaffecting>5%ofhepatocytes),usuallyassociatedwithmild chronicinﬂammation[72,73].

Steatosisresultsfromtheinteractionofmultiplefactors includ-ingage,sexandlifestyle[70,73].Bodyweightmirrorsdietaryhabits andphysicalexerciseandisamajordeterminantofNAFLD.Body weightchanges ofas littleas2–3kgareassociated witheither increasedrisk ofdeveloping NAFLDor itsreversal[74].Finally, variousdrugsmaycontributetoreversingNAFLD[5].

3.2.2. NASH

Steatosisplusmildinﬂammatorychangesand“ballooning” hep-atocytedegenerationdeﬁneNASH,whichaccountsforupto30%of NAFLDcases[73].

ThestrongestpredictorsofNASHare:olderage,malesex,and variousgeneticandmetabolicriskfactors[75–81].

SimplesteatosiscancertainlyprogresstoNASH[82–84].The timetoprogresstocirrhosisfromearly,diseasewouldprobably takeaslongasapproximately57yearsforsteatosiscomparedto24 yearsforNASH[3].Theratesofdiseaseprogressionareslowerthan thoseobservedinHCVinfection[85].

Glitazonesdonotsigniﬁcantlyreversehepatocyteballooning; thisﬁndingsupportstheviewthathepatic/systemicIRplaysakey roleintriggeringtheearlyphasesofNAFLDdevelopmentrather thansustainingitssubsequentprogression[86].

3.2.3. Advancedﬁbrosis/cirrhosis

NASH strongly predicts thedevelopment of advanced fibro-sistogetherwith age,body massindex (BMI),sex, geneticand hormonal/metabolicfactors[87–94].Fibrosis,initsturn,strongly predictsliver-relatedmortalityinNAFLD[95].Statinsseemalsoto exertsomebeneficialeffectonliverfibrosisprogressioninpatients withandwithoutT2DM[96,97].

NASH-cirrhosis histologically exhibits advanced ﬁbrosis and nodularchanges;however,fattychangesmaytypicallydisappear overtime[98].

Obesity, T2DM and steatosis, together withthe presence of carotidatheroscleroticplaquesandincreasedintima-media thick-ness, are strongly associated with advanced ﬁbrosis/cirrhosis [99–101]. In particular, the coexistence of T2DM and steatosis

(5)

stronglypredictsthedevelopmentofclinicallysigniﬁcanthepatic ﬁbrosis[100].

3.2.4. HCC

ComparedtootheretiologiesofHCC,NAFLD-HCCisassociated withalowermale/femaleratioandmayoccurinnon-cirrhotic liv-ers[102–104].NAFLD-HCCisoftendiagnosedlateasaresultof itsescapingsurveillanceprotocols,whichtranslatesintocurtailed chancesforradicaltreatmentand accountsforaworse progno-siscomparedwithHCCcausedbyviralhepatitis[104–106].The PNPLA3I148Mpolymorphism,age,BMI,T2DM,dietaryhabitsand drugsmaybi-directionallymodulatetheriskofdeveloping NAFLD-HCC[92,107–113]andT2DMaggravatesHCCoutcome[114]. 3.3. NaturalhistoryofNAFLD:cardiovascular,arrhythmicand diabetesrisks

Cardiovasculardisease(CVD)istheleading causeof mortal-ity in patients with NAFLD [95,115–117]. A recent systematic reviewandmeta-analysisshowedthatpatientswithNAFLDhad ahigherriskoffatalandnon-fatalCVDeventsthanthose with-outNAFLD[66].Patientswithmore‘severe’NAFLDwerealsomore likely todevelopfatal and non-fatalCVDevents[66].Although theresultsofthisupdatedmeta-analysisproviderobustevidence oftheassociationbetweenNAFLD(andNAFLDseverity)andrisk ofmajor CVDevents, itis importantto highlightthatcausality remainstobefurtherproveninhigh-qualityinterventionstudies. GiventhatCVDcomplicationsfrequentlydictatetheoutcomeof NAFLD,ascreeningofthecardiovascularsystemismandatoryin allpatientswithNAFLD,atleastbydetailedriskfactorassessment [11].

NAFLDisalsolinkedwithsubclinicalmyocardialremodeling anddysfunction(i.e.,functionalandstructuralcardiomyopathy), valvularheartdiseases(i.e.,aortic-valvesclerosisormitralannulus calciﬁcation),andincreasedprevalenceandincidenceof perma-nentatrialﬁbrillation[118–124].Moreover,NAFLDisstronglyand independently associated with heart rate-corrected QT interval prolongation[125,126]andanincreasedprevalenceofventricular arrhythmiason24-hECG-Holtermonitoring[127].Finally, prelim-inaryevidencealsosuggeststhatNAFLDisassociatedwithhigher 1-yearre-hospitalizationratesin patientshospitalizedforacute heartfailure[128].

For many years, NAFLD has been considered as the simple “hepaticmanifestationoftheMetS”[63,129].However,alarge num-berof retrospectiveandprospective observationalstudieshave recentlydemonstratedthatNAFLDisanearlypredictorofanda determinantforthedevelopmentofnew-onsetT2DMandMetS [4,130].Arecentsystematicreviewandmeta-analysisconﬁrmed thatNAFLD(asdiagnosedeitherbyabnormalserumliverenzymes orbyultrasonography)isindeedassociatedwithalmosttwo-fold increasedincidenceratesofbothT2DMandMetSoveramedian 5-year follow-up period [7]. Consistently, recent observational studieshaveshownthatthereisastrongandindependent asso-ciationbetweenimprovementofNAFLDanddecreasedincidence ofT2DM[131,132].Futurecontrolledinterventiontrialsare, there-fore,neededtodeterminewhethertreatingNAFLDleadstoT2DM riskreduction.

3.4. NaturalhistoryofNAFLD:cancerrisk

IndirectevidenceforanassociationofNAFLDwithcancer inci-dence and mortality came from large epidemiological studies, whichhaveclearlydemonstratedthatobesityandT2DMcontribute toincreasedincidenceandmortalityfromseveralcancertypes, includingliverandcolo-rectalcancers[133–135].Consistently, lon-gitudinalstudiesshowedthatNAFLDwasassociatedwithincreased

burdenofhepaticandextra-hepaticcancers[136]andmalignancy wasaleadingcauseofmortalityinpatientswithNAFLD,ranking secondafterCVDmortality[137]andaccountingforuptoonethird ofdeathsinpatientswithNAFLDandT2DM[138].

Primarylivercancerisdeemedtobecausallyassociatedwith NAFLD.HCCincidenceissteadilyincreasingoveryearsin paral-lelwiththeburstof“diabesity”.NAFLDisbecomingamajorcause ofHCC andthesecond mostcommoncause ofHCCin patients listedforlivertransplantation[106,139].Worryingly,thewhole histological spectrum of NAFLD can lead to HCC and as many as 50%of NAFLD-relatedHCC occurin patientswithout cirrho-sisandareoftendetectedlate[104].CumulativeHCCincidence or mortality rates seem tobe lower in NASH-cirrhosisthan in cirrhosis due toother etiologies;however, it is notpossible to clearly gauge theactual HCC risk in thewholeNAFLD popula-tion[140–142].Apartfromadvancedfibrosis/cirrhosis,dataonly point to male sex, older age, alcohol abuse and MetS compo-nentsasestablishedclinicalriskfactorsforHCCdevelopmentin NAFLD[108,143,144].Forthesereasons,exceptforpatientswith NASH-cirrhosis, specificsurveillance/screeningstrategiescannot becurrentlyrecommendedinNAFLDpatients.Mountingevidence alsosuggeststhatMetSandNAFLDareinvolvedinthedevelopment ofothertypesofprimarybenignandmalignantlivertumors,i.e. hepatocellularadenoma[145,146]andintra-hepatic cholangiocar-cinoma[147–150].Ofnote,theinterplaybetweenNAFLDandHCC seemstobemorecomplexthantheclassiclinearequation fibrosis-cirrhosis-dysplasia,andprobablyhepatocellularadenomaplaysa roleinthisprocess[103,151,152].

Emergingevidence alsosuggestsa link betweenNAFLDand someextra-hepaticcancers,typicallythosecloselyassociatedwith “diabesity” [153]. In particular, colorectal adenoma and cancer havebeenassociatedwithNAFLDbyretrospectivestudiesfrom Asia,USAandEurope[154–157].Therearealsoisolatedreports of anincreasedrisk ofothergastrointestinal(esophagus, stom-achandpancreas)andextra-gastrointestinalmalignancies(kidney, prostate,lungandbreast)[136,153,158–160].However,allthese dataarepreliminary,and theirvalidityremainstobeevaluated prospectivelybeforespeciﬁcsurveillance/screeningstrategiesmay beissued.

4. DiagnosisandclinicalapproachtoNAFLD 4.1. Diagnosis

Given thatapproximatelyone third oftheadultEuropeand USApopulationshaveNAFLDandthatinvasivediagnostic tech-niquesarenotapplicableinthisimpressivenumberofindividuals, non-invasivebiomarkersofsteatosisandﬁbrosisarenecessary.As showninFig.2,therecentlypublishedEuropeanclinicalpractice guidelinesforthemanagementofNAFLDhaveproposeda diagnos-ticﬂow-charttoassessandmonitordiseaseseverityinthepresence ofsuspectedNAFLDandmetabolicriskfactors[11].

ThediagnosisofNAFLDshouldrelyonthefollowingcriteria: (i)hepaticsteatosisoneitherimagingorhistology,(ii)no exces-sivealcoholconsumption(athresholdof20g/dayforwomenand 30g/dayformenisconventionallyadopted),and(iii)nocompeting causesofhepaticsteatosis[11].

LiverbiopsyremainsthereferencestandardfordiagnosingNASH andstagingﬁbrosisinpatientswithNAFLD.However,this proce-dureisinvasive,potentiallyrisky,patientunfriendlyandsubjectto samplingerror;therefore,liverbiopsyisnotsuitablefordiagnosis inlargecohortsofindividualsorforpatientmonitoring[11].

Imaging techniques. The gold standard to diagnose hepatic steatosisisthemagneticresonanceimaging(MRI)orspectroscopy (MRS)thatcandetectamountoffataslowas1%[11].However, the cost of MRI limits itsroutine clinical use. Ultrasonography

(6)

Fig.2.Diagnosticflow-charttoassessandmonitordiseaseseverityinthepresence ofsuspectedNAFLDandmetabolicriskfactors.1_Steatosis_biomarkers:_Fatty_Liver Index,SteatoTest,NAFLDfatscore.2_Liver_tests:_ALT,_AST_orGGT.3_Any_increase_in_ALT, ASTorGGT.4_Serum_fibrosis_markers:_NAFLD_Fibrosis_Score,_FIB-4,_Commercial_tests (e.g.,FibroTest,FibroMeterorELFscores).5_Low_risk:_indicative_of_no/mild_fibrosis; medium/highrisk:indicativeofsignificantfibrosisorcirrhosis.

Reprinted,withpermission,fromRef.[11].

remainstherecommendedﬁrst-lineimagingmodalityinclinical practice,butitsaccuracyismuchlowerthanthatofMRI/MRSand itscostisalsolowerthanthatofMRI.Recently,withthe devel-opmentofnewimagingsequencesforMRI,timeofscreeningis decreased(approximately 10–15min/session) andthus thecost canbecomemoreaffordable[161,162].Otherimagingtechniques, e.g.ultrasonography-basedtransientelastography(Fibroscan),2D acousticradiationforceimpulseimagingorMR-elastography,can alsobeusedtodiagnoseliverﬁbrosis[11].Themainlimitationof ultrasonography-basedtransientelastographyinclinicalpractice isitsfailuretoobtainreliableliverstiffnessmeasurements(∼20% ofcases),mainlyinobesepatients,whichdiminishesitsapplication inNAFLD.

Serumbiomarkers. Hepatic steatosisis often associated with mild-to-moderate elevations of serum aminotransferase and gamma-glutamyl transferase (GGT) levels, but, at best, liver enzymesonlyidentifypeoplewhoareatincreasedriskofNAFLD andwho requirefurtherdiagnostictests.However, serumliver enzymelevelsareinaccurateindicatorsforand,therefore,should notbeusedaloneinclinicalpractice[11].Amongthebestscores forsteatosistherearethefattyliverindex(FLI)[163],theNAFLD liverfatscore[164]andtheSteatoTest[165].Theﬁrsttwoscores canbeeasilycalculatedusingserumtriglycerides,GGT,insulinand aminotransferases,BMI,waistcircumferenceandpresenceofMetS orT2DM[166].TheSteatoTestisacommercialkit.

AcommonclinicalconcerninpatientswithNAFLDiswhether theyhavesimplesteatosisorNASHand,moreimportantly,what thestageofhepatic ﬁbrosisisandwhetherthelevelofﬁbrosis (whichdictatesclinicaloutcome)hasincreasedovertime.There

arevarioushepaticfibrosisnon-invasivebiomarkers:theNAFLD fibrosisscore [167],theFIB-4 [168],theFibroTest, the Fibrom-eter,andtheEnhancedLiverFibrosis(ELF)score[165,169].The firsttwoscorescanbecalculatedusingplateletcount,albumin, andaminotransferases.TheFibroTest,FibrometerandELFscores arecommercialtests.Althoughnon-invasivemethodsrequire fur-thervalidation,thevarioustestscouldbeusefulforselectingthose patientswithNAFLDwhowillrequirealiverbiopsy.However,for thediagnosisof NASHbiochemical testsor imagingtechniques cannotdistinguishNASHfromsimplesteatosisand liverbiopsy remainsthereferencestandard[11].

4.2. Clinicalapproach

CVDcomplicationsarecommoninNAFLDpatientsandtherisk ofCVDandpoorCVDoutcomesisparticularlyincreasedinthose NAFLDpatientswithsevereliverdamage[95,117],suggestingthat NAFLDexertsapro-atherogeniceffectviaapro-inﬂammatoryand pro-ﬁbrogenicmilieuthatcharacterizesseveredisease[170].

Onthesegrounds,acarefulassessmentofcardiovascularrisk (CVR)inallpatientswithNAFLDiswarranted.Itshouldincludethe assessmentoftraditionalCVRfactorsandalsoindicesofseverityof liverdisease,(serumsodiumandalbumin,theModelforend-stage liverdisease(MELD)score andtheNAFLDﬁbrosis score[171]). WhetherPNPLA3[172]andTM6SF2[54]genevariants,whichare associatedwithCVDalterationsinNAFLD,mayhelptostratifythe CVRneedsfurtherinvestigation.

Overall,bothhepaticandextra-hepaticriskfactorsshouldbe evaluatedinallpatientswithNAFLD.Namely, besidesthe non-invasiveassessmentofliverdamage(asdescribedabove),acareful searchforcardio-metaboliccomorbiditiesshouldalsoincludethe evaluationoffamilyhistoryforCVD/T2DM,theassessmentofBMI andwaist circumference,themeasurementof plasmalipid val-ues,estimatedglomerularfiltrationrateandalbuminuriaand,in selectedcases,thegeneticassessmentforstoragedisorders,such asthelysosomalacidlipasedeficiency[116,173].Innondiabetic patients,screeningforT2DMbymeasuringfastingorrandomblood glucose or hemoglobinA1c levelsis alsomandatory. Similarly, repeatedassessmentofarterialpressureshouldberecommended tonon-hypertensivepatients.Finally,CVRassessmentshouldrely notonlyontheavailableCVR scoringsystems,suchasthe Fra-minghamor“Progettocuore”riskcharts,butalsoontheassessment oftheseverityofliverdisease[171].In patientsathighCVRor withestablished T2DM,referral tocardiologistsshould alsobe consideredandfurtherdiagnostictestsperformedwhenever indi-cated.In nondiabetic patientsat low CVR, re-assessmentevery 2–3years may besuggested based onclinical judgmentalone. Patientswithadvancedfibrosis/cirrhosisshouldenterappropriate follow-upschedulesaimedatanearlydiagnosisofHCCandportal hypertension.However,ageneralizedHCCscreeningpolicycannot berecommendedinallnon-cirrhoticNAFLDpatientsatpresent, althoughcertainNAFLDindividualsareatanincreasedrisk.

Finally,inpatientswithdyslipidemia,hypertensionorT2DM, thecarefulcontrolofthesecardiometabolicdisorderswillnotonly reducetheNAFLD-associatedmetabolicandCVR,butalsoslowliver diseaseprogression.Toachievethisgoalstatins,insulin-sensitizers, incretin mimetics and ACE-inhibitors/sartans, if necessary, can safelybeused.

5. TreatmentofNAFLD/NASH

5.1. Nutritionalaspectsandlifestylechanges

Alltherecommendations fromthemostimportantscientiﬁc societies indicatethat body weightreduction obtainedthrough

(7)

lifestylechangesisthemosteffectivetreatmentforNAFLD.Alarge bodyofevidenceindicatesthatunhealthylife-styleisarelevant riskfactorforNAFLD/NASH[17,174].Accordingly,lifestylechanges focusingonweightloss,physicalactivityandhealthydietrepresent thebesttherapeuticapproach[175].

Abodyweightreductionof7–10%obtainedwithenergy restric-tionand/orregularphysicalactivityisassociatedwithhistological improvement/resolutionofsteatosisandinﬂammationand regres-sionofﬁbrosis[176,177].

Energyrestrictionshouldbeobtainedwithalow-calorie(about 1200–1600kcal/day)low-fatlow-carbohydratediet[178].Thediet compositioninNAFLDshouldconsistof<30%fat,notablywithno morethan10%ofsaturatedfattyacid,andrelativelylowin carbo-hydrates,∼50%oftotalkcal,notablybyreducinghigh-sugarfood [179].

As for physical activity, both aerobic activities (150–200min/week of moderate intensity aerobic exercise in aboutfoursessions,i.e.brisk walkingand/orstationerycycling) and resistance training effectively reduce hepatic fat content. Trainingshould be tailoredbased on patients’ preferences and maintainedlong-term[177,180,181].

Changesofdietcomposition,especiallyfornon-obesepatients, arearealisticandfeasibleapproachevenifbasedonlessconsistent evidence[182].Thegoldstandardoptionisaﬁber-richdiet rela-tivelylowin carbohydrates(complexonesshouldbepreferred) andfatcontent,withtheavoidanceoffructose[179],amoderate useof coffeeandadherencetoaMediterraneandietarypattern [183,184].Indeed,amongalltheproposeddiets,theMediterranean dietappearsasthemosteffectivedietaryoptionfor inducinga weightlosstogetherwithbeneﬁcialeffectsonallcardio-metabolic riskfactorsassociatedwithNAFLD.

5.2. NewdrugsforNASH

DrugtherapyshouldbereservedforNASHpatientswhoareat maximalriskfordiseaseprogression.Todate,thereareveryfew high-quality,randomized,blinded,adequatelypowered,controlled studiesofsufﬁcientdurationandwithadequatehistological

out-comes.Asafact,currentlytherearenoapprovedavailabledrugs forthetreatmentofNASH.Moreover,drugtreatmentofNASHhas beenfocusedtowardreducingsteatosis,necro-inﬂammationand ﬁbrosis.

Given that NASH is closely related to T2DM, hypoglycemic drugshavebeenlargelyevaluated.InpatientswithT2DM,incretin mimetics,whichactbyagonizingglucagon-likepeptide1 (GLP-1)receptor,haveproveneffectiveinreducinghepatic/systemicIR, serumliverenzymesandliverfatcontent.TheLEANtrialperformed on52obesepatientsshowedthatliraglutidewasabletoresolve histologicfeaturesofNASHin39%ofthetreatedpatients[185]. Theserespondersalsoregisteredameanweightlossof2.1kg rais-inguncertaintyastowhetherthebeneficialhepaticeffectwasdue toliraglutideperseorcombinedwithweightloss.Further long-termstudieswithliraglutideareneededtoconfirmitsefficacyin patientswithNASH.

Otherinsulin-sensitizers,suchasPPAR-␥agonistshave been evaluatedinNASH[186].Thebestlevelofevidenceisavailablefor pioglitazoneinbiopsy-provenNASHpatients[187].PPARs modula-tionresultsintranscriptionalregulationofseveralgenesinvolved inmetabolicpathways[188]andthemodulationofadual PPAR-␣/␦agonistthroughelafibranoriseffectiveinimprovingplasma lipidprofile,andhepatic/systemicIR[189].InthephaseIIbGOLDEN trial,elafibranorwastestedwithaprimaryhistologicalend-point (NASHresolutionwithoutfibrosisworsening),whichwasachieved in 23% and 21% of patients treated with either 80mg/day or 120mg/day,respectively[190].Post-hocanalysisshowedaclear benefitinpatientswithmoreseveredisease,promptingaphaseIII trial.

AnothernuclearreceptortobetargetedtoobtainreducedIRand liverinjury,isFXR,anintracellularbileacidsreceptor[191]capable ofinhibitingbileacidssynthesis,reducingdenovolipogenesisand steatosisbyimprovingIR.IntheFLINTtrial,treatmentwith obeti-cholicacid,apotentFXRactivator,achievedaprimaryend-point ofimprovingthenecro-inflammationwithoutworseningof fibro-sisin46%ofthetreatedpatients.Moreover,comparedtoplacebo, theNASHresolutionwasobtainedin22%oftreatedpatients[192]. However,efficacyandlong-termsafety(e.g.,pruritusandincreased

Table1

OngoingclinicaltrialsforNASHinadultsubjects.

Drug Clinicaltrial number

Moleculartarget Routeof administration

Phaseof clinicaltrial

Histological end-points Drugstargetingmetabolichomeostasis

Elaﬁbranor NCT02704403 Peroxisome

proliferator-activatorreceptor ␣/␦agonist

Oral,once-daily Phase3 Yes

Dapagliﬂozin NCT02696941 Sodium-glucoseco-transporter 2inhibitor

Oral,once-daily Pilot No Empagliﬂozin NCT02964715

and NCT02637973

Sodium-glucoseco-transporter 2inhibitor

Oral,once-daily Pilot Yes

Tofogliﬂozin NCT02649465 Sodium-glucoseco-transporter 2inhibitor

Oral,once-daily Pilot Yes Semaglutide NCT02970942 Glucagon-likepeptide1

analogue

Subcutaneous injection, once-daily

Phase2 Yes

Liraglutide NCT02721888 Glucagon-likepeptide1 analogue Subcutaneous injection, once-daily Pilot No MSDC-0602K NCT02784444 Mitochondrialtargetof thiazolidinedionesmodulator

Oral,once-daily Phase2 Yes Aramchol NCT02279524 StearoylCoAdesaturase

inhibitor

Oral,once-daily Phase2b Yes,assecondary outcomes GS-0976 NCT02856555 Acetyl-CoAcarboxylase

inhibitor

Oral,once-daily Phase2 No TVB-2640 NCT02948569 Fattyacidsynthaseinhibitor Oral,once-daily Phase1/2b _No

ARI-3037MO NCT02574325 Niacinanalogue Oral,twice-daily Phase2 No

AZD4076 NCT02826525 MicroRNA103/107antagonist Subcutaneous

injection

(8)

Table1(Continued)

Drug Clinicaltrial number

Moleculartarget Routeof administration

Phaseof clinicaltrial

Histological end-points MGL-3196 NCT02912260 Liver-targetedselectiveagonist

forthethyroidhormone receptor-␤

Oral,once-daily Phase2 Yes,assecondary outcomes VK2809 NCT02927184 Liver-targetedselective

eagonistforthethyroid hormonereceptor-␤

Oral Phase2 No

Obeticholicacid NCT02548351 SyntheticbileacidfarnesoidX receptoragonist

Oral,once-daily Phase3 Yes LJN452 NCT02855164 NonsteroidalfarnesoidX

receptoragonist

Oral Phase2 No

GS-9674 NCT02854605 NonsteroidalfarnesoidX receptoragonist

Oral,once-daily Phase2 No EDP-305 NCT02918929 NonsteroidalfarnesoidX

receptoragonist

Oral,once-daily Phase1 No BMS-986036 NCT02413372 Fibroblastgrowthfactor21

analogue

Subcutaneous injection, once-daily

Phase2 No

NGM282 NCT02443116 Fibroblastgrowthfactor19 analogue Subcutaneous injection, once-daily Phase2 No Volixibat (SHP626)

NCT02787304 Ilealapicalsodium-dependent bileacidtransporterinhibitor

Oral,once-daily Phase2 Yes MT-3995 NCT02923154 Nonsteroidalmineralcorticoid

receptorantagonist Oral Phase2 No Testosterone undecanoate NCT01919294 Hypogonadism Intramuscolar injection Pilot Yes

Somatropin NCT02217345 Growthhormone Injection, once-daily

Pilotb _No

Tesamorelin NCT02196831 Syntheticanalogueofgrowth

hormone-releasinghormone Subcutaneous injection, once-daily Phase2(HIV only) Yes,assecondary outcomes Metreleptin NCT01679197 and NCT02654977 Recombinant-methionyl humanleptin Subcutaneous injection Phase2(pts withpartial lipodystrophy) Yes

Drugstargetingoxidativestress,inﬂammationandapoptosis

VitaminE NCT01792115

and NCT02962297

Antioxidant Oral Phase2-3b _Yes

Metadoxine NCT02541045 Antioxidant Oral,twice-daily Phase3b _Yes

Emricasan NCT02686762

and NCT02960204

Caspaseinhibitor Oral,twice-daily Phase2a _Yes

GS-4997 NCT02781584

and NCT02466516

Apoptosissignal-regulating kinase1inhibitor

Oral,once-daily Phase2 No

Cenicriviroc NCT02330549

and NCT02217475

DualC-Cchemokinereceptor types2and5antagonist

Oral,once-daily Phase2 Yes

JKB-121 NCT02442687 Toll-likereceptor4antagonist Oral,twice-daily Phase2 No

Amlexanox NCT01975935

and NCT01842282

TANK-bindingkinase1andI␬B kinase␧inhibitor

Oral,thrice-daily Phase2 No

CF102 NCT02927314 A3adenosinereceptoragonist Oral,twice-daily Phase2 No

MN-001 (tipelukast)

NCT02681055 Multipletargets(leukotriene

receptorantagonism, inhibitionof

phosphodiesterases3and4, inhibitionof5-lipoxygenase)

Oral Phase2 No

Drugstargetingﬁbrosis

GR-MD-02 NCT02421094 Galectin-3inhibitor Intravenous

infusion

Phase2a _No

Simtuzumab NCT01672866

and NCT01672879

Lysyloxidase-like2antibody Subcutaneous

injec-tion/intravenous infusion

Phase2a _Yes

Otherdrugs

Solithromycin NCT02510599 Antibiotic Oral,once-daily Phase2 Yes

IMM-124E NCT02316717 IgG-richbovinecolostrum Oral,thrice-daily Phase2 No

Fecalmicrobiota transplantation

NCT02469272 and NCT02868164

Gutmicrobiota Duodenalinfusion Pilota _No

FromClinicalTrials.govaccessedon6thDecember,2016[InterventionalStudieswithDrugs/BiologicalAgents;OpenStudies(recruiting;notyetrecruiting;expandedaccess) orClosedStudies,butActive,NotRecruiting,withoutResults;NonAlcoholicFattyLiverDisease/NonAlcoholicSteatohepatitis].

a_Trials_including_patients_with_cirrhosis. b _Trials_not_including_patients_with_type₂_diabetes.

(9)

LDL-cholesterollevels)associatedwiththeuseofthisdrugneedto beaddressed.AndaphaseIIIstudyisnowongoingtothisend.

AnothertherapeuticoptionforNAFLDmaybetodecrease oxida-tivestressbyadministrationofanantioxidant,suchasvitaminE. InthePIVENStrial[187],in247nondiabeticNASHadults, vita-minEtreatment(800U/dayfor 96weeks),wasassociated with significantimprovementsinserumliverenzymesandNASH histo-logicalfeatures(steatosis,inflammationandballooning)compared toplacebo.However,beforevitaminEcanberecommendedforthe treatmentofNAFLD,furtherstudiesarerequiredtosupportefficacy andsafetyofthisfat-solublevitamin.

Attoday,onlyelaﬁbranorand obeticholicacidare recruiting forphaseIIItrials;resultsareexpectedin2020.Moreover,several otherphaseIItrialsareinprogresstoevaluatetheeffectofnewer drugsondifferentpathogenicpathwaysinvolvedinNASH.Table1 recapitulatesongoingclinicaltrialsforNASHinadultsubjects.

Promisingresultsareexpectedfornoveldrugs,whichimprove metabolicdisturbancessuchasAramchol(phaseIItrial),a conju-gateofcholicandarachidicacid,whichpartiallyreducestheactivity of stearoyl-CoA desaturase and de novo lipogenesis [193,194]. Emricasan [195], an oral pan-caspase inhibitor acting on TNF-alphadriveninflammation,seemstobeeffectiveinreducingliver damagebymodulatingapoptosis.Cenicriviroc[196],anoral antag-onistof chemokineCCR2/CCR5,hasrecentlyclosedthephaseII trialand resultsareexpected.Thepossibilityofreverting fibro-sisand cirrhosis has alsobeen exploredby using simtuzumab, a monoclonal antibodyagainst lysyloxidase-like-2, which is an extra-cellular amine oxidase involved in the post-translational modificationofcollagensandelastin intheextracellularmatrix [197].

6. Conclusionsandperspectives

AgreatdealofnewknowledgeinthephysiopathologyofNAFLD hasbeenaccumulatingoverthelastdecade,revealingthe complex-ityofthemechanismsinvolvedinthedevelopmentandprogression ofthiscondition.Themostrecentguidelines/expertopinionsfor NAFLDmanagementpromptanew“systemsmedicine”approach totheinterplaysbetweenbrainand nervoussystem,endocrine system,digestivesystem(gut,liverandmicrobiota)andimmune system[11,198].Newconceptsforpatientstratiﬁcationareneeded toidentifydifferentclinicalprototypeswithintheindistinct phe-notypeoftheMetS[11,198].

The emerging and rapidly growing field of the molecular imagingwilllikelybeprovidinganextraordinarynewopportunity for non-invasive studies of in vivo physiopathology of NAFLD overcomingtheinherentlimitationsofliverbiopsy[161,199,200]. Suchlimitationshavesofarhamperedtheprogressofthe knowl-edgein key fieldssuchastheimpactand roleof autophagyin thephysiopathologyofNAFLD[201].Studiescombiningthenew imagingtechniqueswithliverandblood metabolomicsandthe analysisoftheinterplaybetweenspecificgenesandtheepigenetic factorsconditioningtheirexpression,willpaveanattractivenew waytotargetingthedynamicsofpathogenicprocessesinvolvedin NAFLDandNASH.Theresultsoftheseongoingstudieswillidentify novel risk factors, new diagnostic and prognostic biomarkers andtherapy targetsfora betterprevention,outcomeprediction and personalized treatment of NAFLD/NASH. In the meantime, tohalt/reducethe tide ofNAFLD, referenceshould bemadeto standardprinciplesofgoodclinicalpracticeforthemanagement ofthiscondition[11,15,202,203],whichisexpectedtoimpacton thefutureglobalburdenofdiseaseworldwide.

Conﬂictofinterest

None todeclare:Stefano Bellentani,Amedeo Lonardo,Gianluca SvegliatiBaroni,GiovanniTargher,LucaValentiMauroBernardi:

CSLBehringGmbH(consultancyandspeaker);BaxterHealthcare SA(consultancyandspeaker);PPTAEurope,GileadSciences, Abb-VieItalia(speaker).

FerruccioBonino:AdivsoryBoardsand/orSpeakersBureaufor Abbott/Abbvie,BaldacciLaboratories,BMS,Fujirebio,GSK,Gilead, MSD,NovartisandRocheElisabettaBugianesi:Consultantfor Gen-ﬁt,IBSA,Innova,BoehrigerIngelheim,InterceptAlessandroCasini: ProbiosSrL:researchcontractAmaliaGastaldelli:Consultantfor Roche,Eli-Lilly,Menarini,Sanoﬁandhasreceivedresearchsupport fromAmylin-BMSAZ.

Giulio Marchesini: Consultant and/or speaker for: Eli Lilly, AstraZeneca,Novartis,IBSA.Clinicalstudies:NovoNordisk,Sanoﬁ, Boehringer Ingelheim, GlaxoSmith Kline, Janssen, Gilead, Gen-ﬁt.

FabioMarra:Abbvie,AstraZeneca,Bayer,Menarini:consultant fees.Gilead,Bayer,ViiVHealthcare:speakerhonoraria.

LucaMiele:AdvisoryBoard:Synageva,MyGenomics,IBSA,MSD, Boehringer-Ingelheim, BMS Speaker fee: Rottapharm Madaus, MEDAFilomenaMorisco:NathuraandIBILorenzini:research con-tract.BMSandAbbvie:speakerfeeSalvoPetta:AdvisoryBoards and/orSpeakerforGilead,AbbVie,Janssen,Bristol-MyersSquibb, MerckSharpandDohmeFabioPiscaglia:Bayer(speakerfeeand advisoryboard).Bracco(speakerfee).Esaote(researchcontract). Meda(speakerfee).Eisai(advisoryboard).

AppendixA

b _Division_of_Internal_Medicine,_NOCSAE,_Department_of

Biomed-ical, Metabolic and Neural Sciences, University of Modena and ReggioEmilia,Modena,Italy

c _Section_of_{Endocrinology,}_Diabetes_and_Metabolism,_Department

ofMedicine,UniversityofVerona,Italy

d _Department _of _Medical _and _Surgical _Sciences, _University _of

Bologna,“AlmaMaterStudiorum”,Bologna,Italy

e _University_of_Pittsburgh_Medical_Center_Institute_for_Health,

Chi-anciano Termeand “Fondazione Italiana Fegato”, AREA Science Park,CampusBasovizza,Trieste,Italy

f _{Gastroenterology}_and _Hepatology,_Department_of _Medical

Sci-ences,Hospital“Città dellaSaluteedellaScienza”,Universityof Turin,Turin,Italy

g _Department _of _Experimental_and _Clinical _Medicine,_School_of

Medicine,UniversityofFlorence,Italy

h_{Cardiometabolic}_Risk_Laboratory,_Institute_of_Clinical_Physiology,

CNR,Pisa,Italy

i _Unit_of_Metabolic_Diseases_and_Clinical_Dietetics,_University_of

Bologna,“AlmaMaterStudiorum”,Bologna,Italy

j _Department_of_Experimental_and_Clinical_Medicine,_University_of

Florence,Italy

k_Internal_Medicine_and_{Gastroenterology}_Area,_University_Policlinic

Foundation“A.Gemelli”,CatholicUniversityofRome,Italy

l _{Gastroenterology} _Unit, _Department _of _Clinical _Medicine _and

Surgery,UniversityofNaples“FedericoII”,Italy

m _Section_of_{Gastroenterology,}_Di.Bi.M.I.S_Policlinic_“Paolo

Giac-cone”Hospital,UniversityofPalermo,Italy

n _Unit_of_Internal_Medicine,_Department_of_Medical_and_Surgical

Sciences,UniversityofBologna,“AlmaMaterStudiorum”,Bologna, Italy

o _Department_of_{Gastroenterology,}_{“Politecnica}_delle_Marche”

Uni-versity,Ancona,Italy

p _Internal _Medicine, _IRCCS _Foundation _“Ca’ _Granda” _Hospital

MaggiorePoliclinico,DepartmentofPathophysiologyand Trans-plantation,UniversityofMilan,Italy

q _{Gastroenteroloy}_and_Hepatology_Service_“Clinica_Santa_Chiara”,

(10)

References

[1]MarchesiniG,PettaS,DalleGraveR.Diet,weightloss,andliverhealthin nonalcoholicfattyliverdisease:pathophysiology,evidence,andpractice. Hepatology2016;63:2032–43.

[2]ValentiL,RomeoS.DestinedtodevelopNAFLD?Thepredictorsoffattyliver frombirthtoadulthood.JournalofHepatology2016;65:668–70.

[3]SinghS,AllenAM,WangZ,etal.Fibrosisprogressioninnonalcoholicfatty livervsnonalcoholicsteatohepatitis:asystematicreviewandmeta-analysis ofpaired-biopsystudies.ClinicalGastroenterologyandHepatology:The Ofﬁ-cialClinicalPracticeJournaloftheAmericanGastroenterologicalAssociation 2015;13:643–54[e1–9;quize39–40].

[4]TargherG,MarchesiniG,ByrneCD.Riskoftype2diabetesinpatientswith non-alcoholicfattyliverdisease:causalassociationorepiphenomenon? Dia-betes&Metabolism2016;42:142–56.

[5]SinghS,KheraR,AllenAM,etal.Comparativeeffectivenessof pharmacolog-icalinterventionsfornonalcoholicsteatohepatitis:asystematicreviewand networkmeta-analysis.Hepatology2015;62:1417–32.

[6]RinellaME,SanyalAJ.ManagementofNAFLD:astage-basedapproach.Nature ReviewsGastroenterology&Hepatology2016;13:196–205.

[7]BallestriS,ZonaS,TargherG,etal.Nonalcoholicfattyliverdiseaseis asso-ciatedwithanalmosttwofoldincreasedriskofincidenttype2diabetesand metabolicsyndrome.Evidencefromasystematicreviewandmeta-analysis. JournalofGastroenterologyandHepatology2016;31:936–44.

[8]NobiliV,MarcelliniM,MarchesiniG,etal.Intrauterinegrowthretardation, insulinresistance,andnonalcoholicfattyliverdiseaseinchildren.Diabetes Care2007;30:2638–40.

[9]WangT,HuangT,LiY,etal.Lowbirthweightandriskoftype2diabetes:a Mendelianrandomisationstudy.Diabetologia2016;59:1920–7.

[10]RatziuV,MarchesiniG.Whenthejourneyfromobesitytocirrhosistakesan earlystart.JournalofHepatology2016;65:249–51.

[11]EuropeanAssociationfortheStudyoftheLiver,EuropeanAssociationforthe StudyofDiabetes,EuropeanAssociationfortheStudyofObesity. EASL-EASD-EASOClinicalPracticeGuidelinesforthemanagementofnon-alcoholicfatty liverdisease.JournalofHepatology2016;64:1388–402.

[12]BugianesiE.EASL-EASD-EASOClinicalPracticeGuidelinesforthe manage-mentofnon-alcoholicfattyliverdisease:diseasemongeringorcalltoaction? Diabetologia2016;59:1145–7.

[13]Wong VW,ChalasaniN.Notroutinescreening,butvigilanceforchronic liver disease in patients with type 2 diabetes. Journal of Hepatology 2016;64:1211–3.

[14]WildSH,MorlingJR,McAllisterDA,etal.Type2diabetesandriskof hos-pitaladmissionordeathforchronicliverdiseases.JournalofHepatology 2016;64:1358–64.

[15]LoriaP,AdinolﬁLE,BellentaniS,etal.Practiceguidelinesforthediagnosisand managementofnonalcoholicfattyliverdisease.AdecaloguefromtheItalian AssociationfortheStudyoftheLiver(AISF)ExpertCommittee.Digestiveand LiverDisease:OfﬁcialJournaloftheItalianSocietyofGastroenterologyand theItalianAssociationfortheStudyoftheLiver2010;42:272–82.

[16]BarreraF,GeorgeJ.Theroleofdietandnutritionalinterventionforthe man-agementofpatientswithNAFLD.ClinicsinLiverDisease2014;18:91–112.

[17]AbdelmalekMF,SuzukiA,GuyC,etal.Increasedfructoseconsumptionis asso-ciatedwithﬁbrosisseverityinpatientswithnonalcoholicfattyliverdisease. Hepatology2010;51:1961–71.

[18]BugianesiE,McCulloughAJ,MarchesiniG.Insulinresistance:ametabolic pathwaytochronicliverdisease.Hepatology2005;42:987–1000.

[19]Shulman GI. Ectopic fat in insulin resistance, dyslipidemia, and car-diometabolic disease.The New EnglandJournal of Medicine 2014;371: 1131–41.

[20]DonnellyKL,SmithCI,SchwarzenbergSJ,etal.Sourcesoffattyacidsstored inliverandsecretedvialipoproteinsinpatientswithnonalcoholicfattyliver disease.TheJournalofClinicalInvestigation2005;115:1343–51.

[21]Seppala-LindroosA,VehkavaaraS,HakkinenAM,etal.Fataccumulationinthe liverisassociatedwithdefectsininsulinsuppressionofglucoseproduction andserumfreefattyacidsindependentofobesityinnormalmen.TheJournal ofClinicalEndocrinologyandMetabolism2002;87:3023–8.

[22]MarraF,GastaldelliA,SvegliatiBaroniG,etal.Molecularbasisand mech-anismsofprogressionofnon-alcoholicsteatohepatitis.TrendsinMolecular Medicine2008;14:72–81.

[23]PolyzosSA,ToulisKA,GoulisDG,etal.Serumtotaladiponectinin nonalco-holicfattyliverdisease:asystematicreviewandmeta-analysis.Metabolism: ClinicalandExperimental2011;60:313–26.

[24]CaligiuriA,GentiliniA,MarraF.MolecularpathogenesisofNASH. Interna-tionalJournalofMolecularSciences2016;17.

[25]Neuschwander-TetriBA.Hepaticlipotoxicityandthepathogenesisof non-alcoholic steatohepatitis: the central role of nontriglyceride fatty acid metabolites.Hepatology2010;52:774–88.

[26]Dongiovanni P, Valenti L. Genetics of nonalcoholic fatty liver disease. Metabolism:ClinicalandExperimental2016;65:1026–37.

[27]LeamyAK,EgnatchikRA,YoungJD.Molecularmechanismsandtheroleof saturatedfattyacidsintheprogressionofnon-alcoholicfattyliverdisease. ProgressinLipidResearch2013;52:165–74.

[28]GautheronJ,VucurM,ReisingerF,etal.ApositivefeedbackloopbetweenRIP3 andJNKcontrolsnon-alcoholicsteatohepatitis.EMBOMolecularMedicine 2014;6:1062–74.

[29]MalhiH,KaufmanRJ.Endoplasmicreticulumstressinliverdisease.Journal ofHepatology2011;54:795–809.

[30]CannitoS,PaternostroC,BuslettaC,etal.Hypoxia,hypoxia-inducible fac-torsandﬁbrogenesisinchronicliverdiseases.HistologyandHistopathology 2014;29:33–44.

[31]DixonLJ,BarnesM,TangH,etal.Kupffercellsintheliver.Comprehensive Physiology2013;3:785–97.

[32]RohYS,SekiE.Toll-likereceptorsinalcoholicliverdisease,non-alcoholic steatohepatitisandcarcinogenesis.JournalofGastroenterologyand Hepatol-ogy2013;28(Suppl.1):38–42.

[33]GuoJ,FriedmanSL.Toll-likereceptor4signalinginliverinjuryandhepatic ﬁbrogenesis.Fibrogenesis&TissueRepair2010;3:21.

[34]GanLT,VanRooyenDM,KoinaME,etal.Hepatocytefreecholesterol lipo-toxicityresultsfromJNK1-mediatedmitochondrialinjuryandisHMGB1and TLR4-dependent.JournalofHepatology2014;61:1376–84.

[35]WreeA,MarraF.Theinﬂammasomeinliverdisease.JournalofHepatology 2016;65:1055–6.

[36]Henao-MejiaJ,ElinavE,JinC,etal.Inﬂammasome-mediateddysbiosis regu-latesprogressionofNAFLDandobesity.Nature2012;482:179–85.

[37]FuchsCD,TraussniggSA,TraunerM.Nuclearreceptormodulationforthe treatmentofnonalcoholicfatty liverdisease.Seminarsin LiverDisease 2016;36:69–86.

[38]Marra F, Bertolani C. Adipokines in liver diseases. Hepatology 2009;50:957–69.

[39]NeishAS.Microbesingastrointestinalhealthanddisease.Gastroenterology 2009;136:65–80.

[40]ClementeJC,UrsellLK,ParfreyLW,etal.Theimpactofthegutmicrobiotaon humanhealth:anintegrativeview.Cell2012;148:1258–70.

[41]SommerF,BackhedF.Thegutmicrobiota–mastersofhostdevelopmentand physiology.NatureReviewsMicrobiology2013;11:227–38.

[42]MehalWZ.TheGordianKnotofdysbiosis,obesityandNAFLD.NatureReviews Gastroenterology&Hepatology2013;10:637–44.

[43]TilgH,KaserA.Gutmicrobiome,obesity,andmetabolicdysfunction.The JournalofClinicalInvestigation2011;121:2126–32.

[44]TremaroliV,BackhedF.Functionalinteractionsbetweenthegutmicrobiota andhostmetabolism.Nature2012;489:242–9.

[45]TurnbaughPJ,LeyRE,MahowaldMA,etal.Anobesity-associatedgut micro-biomewithincreasedcapacityforenergyharvest.Nature2006;444:1027–31.

[46]CaniPD,BibiloniR,KnaufC,etal.Changesingutmicrobiotacontrolmetabolic endotoxemia-inducedinﬂammationinhigh-fatdiet-inducedobesityand dia-betesinmice.Diabetes2008;57:1470–81.

[47]PendyalaS,WalkerJM,HoltPR.Ahigh-fatdietisassociatedwithendotoxemia thatoriginatesfromthegut.Gastroenterology2012;142:1100–1,e2.

[48]DeMinicisS,DayC,Svegliati-BaroniG.FromNAFLDtoNASHandHCC: patho-geneticmechanismsandtherapeuticinsights.CurrentPharmaceuticalDesign 2013;19:5239–49.

[49]MiuraK,OhnishiH.Roleofgutmicrobiotaandtoll-likereceptorsin nonalco-holicfattyliverdisease.WorldJournalofGastroenterology2014;20:7381–91.

[50]De Minicis S,Rychlicki C, Agostinelli L, etal. Dysbiosis contributes to ﬁbrogenesis in the course of chronic liverinjury in mice. Hepatology 2014;59:1738–49.

[51]RomeoS,KozlitinaJ,XingC,etal.GeneticvariationinPNPLA3confers suscep-tibilitytononalcoholicfattyliverdisease.NatureGenetics2008;40:1461–5.

[52]DongiovanniP,DonatiB,FaresR,etal.PNPLA3I148Mpolymorphismand pro-gressiveliverdisease.WorldJournalofGastroenterology2013;19:6969–78.

[53]KozlitinaJ,SmagrisE,StenderS,etal.Exome-wideassociationstudy iden-tiﬁesaTM6SF2variantthatconferssusceptibilitytononalcoholicfattyliver disease.NatureGenetics2014;46:352–6.

[54]DongiovanniP,PettaS,MaglioC,etal.Transmembrane6superfamilymember 2genevariantdisentanglesnonalcoholicsteatohepatitisfromcardiovascular disease.Hepatology2015;61:506–14.

[55]MancinaRM,Dongiovanni P, PettaS,etal. TheMBOAT7-TMC4 variant rs641738increasesriskofnonalcoholicfattyliverdiseaseinindividualsof europeandescent.Gastroenterology2016;150:1219–30,e6.

[56]SpeliotesEK,Yerges-ArmstrongLM,WuJ,etal.Genome-wideassociation analysisidentiﬁesvariants associated withnonalcoholic fatty liver dis-easethathavedistincteffectsonmetabolictraits.PLoSGenetics2011;7: e1001324.

[57]DiFilippoM,MoulinP,RoyP,etal.HomozygousMTTPandAPOBmutations mayleadtohepaticsteatosisandﬁbrosisdespitemetabolicdifferencesin congenitalhypocholesterolemia.JournalofHepatology2014;61:891–902.

[58]YounossiZM,KoenigAB,AbdelatifD,etal.Globalepidemiologyof nonalco-holicfattyliverdisease-meta-analyticassessmentofprevalence,incidence, andoutcomes.Hepatology2016;64:73–84.

[59]LonardoA,BelliniM,TartoniP,etal.Thebrightliversyndrome.Prevalenceand determinantsofabrightliverechopattern.ItalianJournalofGastroenterology andHepatology1997;29:351–6.

[60]BedogniG,MiglioliL,MasuttiF,etal.Prevalenceofandriskfactorsfor nonal-coholicfattyliverdisease:theDionysosnutritionandliverstudy.Hepatology 2005;42:44–52.

[61]BellentaniS,BedogniG,MiglioliL,etal.Theepidemiologyoffattyliver. Euro-peanJournalofGastroenterology&Hepatology2004;16:1087–93.

[62]Bellentani S,SaccoccioG, Masutti F, et al. Prevalence ofand risk fac-torsforhepaticsteatosisinNorthernItaly.AnnalsofInternal Medicine 2000;132:112–7.

(11)

[63]LonardoA,BellentaniS,etal.Epidemiologicalmodiﬁersofnon-alcoholicfatty liverdisease:focusonhigh-riskgroups.DigestiveandLiverDisease:Ofﬁcial JournaloftheItalianSocietyofGastroenterologyandtheItalianAssociation fortheStudyoftheLiver2015;47:997–1006.

[64]BrowningJD,SzczepaniakLS,DobbinsR,etal.Prevalenceofhepaticsteatosis inanurbanpopulationintheUnitedStates:impactofethnicity.Hepatology 2004;40:1387–95.

[65]BrunoS,MaisonneuveP,CastellanaP,etal.Incidenceandriskfactorsfor non-alcoholicsteatohepatitis:prospectivestudyof5408womenenrolledin Italiantamoxifenchemopreventiontrial.BMJ2005;330:932.

[66]TargherG,ByrneCD,LonardoA,etal.Non-alcoholicfattyliverdiseaseand riskofincidentcardiovasculardisease:ameta-analysis.JournalofHepatology 2016;65:589–600.

[67]YounossiZM,StepanovaM,NegroF,etal.Nonalcoholicfattyliverdiseasein leanindividualsintheUnitedStates.Medicine2012;91:319–27.

[68]BertolottiM,LonardoA,MussiC,etal.Nonalcoholicfattyliverdiseaseand aging:epidemiologyto management.WorldJournalofGastroenterology 2014;20:14185–204.

[69]MiddletonJP,WienerRC,BarnesBH,etal.Clinicalfeaturesofpediatric non-alcoholicfattyliverdisease:aneedforincreasedawarenessandaconsensus forscreening.ClinicalPediatrics2014;53:1318–25.

[70]BallestriS,NascimbeniF,RomagnoliD,etal.Theroleofnuclearreceptorsin thepathophysiology,naturalcourse,anddrugtreatmentofNAFLDinhumans. AdvancesinTherapy2016;33:291–319.

[71]PetajaEM,Yki-JarvinenH.Deﬁnitionsofnormalliverfatandtheassociation ofinsulinsensitivitywithacquiredandgeneticNAFLD—asystematicreview. InternationalJournalofMolecularSciences2016;17.

[72]KubesP,MehalWZ.Sterile inﬂammationinthe liver.Gastroenterology 2012;143:1158–72.

[73]RinellaME. Nonalcoholicfattyliverdisease:asystematic review.JAMA 2015;313:2263–73.

[74]Zelber-SagiS,LotanR,ShlomaiA,etal.Predictorsforincidenceandremission ofNAFLDinthegeneralpopulationduringaseven-yearprospective follow-up.JournalofHepatology2012;56:1145–51.

[75]DixonJB,BhathalPS,O’BrienPE.Nonalcoholicfattyliverdisease: predic-torsofnonalcoholicsteatohepatitisandliverﬁbrosisintheseverelyobese. Gastroenterology2001;121:91–100.

[76]CarulliL,CanediI,RondinellaS,etal.Geneticpolymorphismsinnon-alcoholic fattyliverdisease:interleukin-6-174G/Cpolymorphismisassociatedwith non-alcoholicsteatohepatitis.DigestiveandLiverDisease:OfﬁcialJournal oftheItalianSocietyofGastroenterologyandtheItalianAssociationforthe StudyoftheLiver2009;41:823–8.

[77]PettaS,CammaC,CabibiD,etal.Hyperuricemiaisassociatedwithhistological liverdamageinpatientswithnon-alcoholicfattyliverdisease.Alimentary Pharmacology&Therapeutics2011;34:757–66.

[78]CarulliL,BallestriS,LonardoA,etal.Isnonalcoholicsteatohepatitis associ-atedwithahigh-though-normalthyroidstimulatinghormonelevelandlower cholesterollevels.InternalandEmergencyMedicine2013;8:297–305.

[79]RehaJL,LeeS,HofmannLJ.Prevalenceandpredictorsofnonalcoholic steato-hepatitisinobesepatientsundergoingbariatricsurgery:aDepartmentof Defenseexperience.TheAmericanSurgeon2014;80:595–9.

[80]DasarathyJ,PeriyalwarP,AllampatiS,etal.HypovitaminosisDisassociated withincreasedwholebodyfatmassandgreaterseverityofnon-alcoholic fattyliverdisease.LiverInternational:OfﬁcialJournaloftheInternational AssociationfortheStudyoftheLiver2014;34:e118–27.

[81]BallestriS,NascimbeniF,RomagnoliD,etal.Theindependentpredictorsof non-alcoholicsteatohepatitisanditsindividualhistologicalfeatures:insulin resistance,serumuricacid,metabolicsyndrome,alanineaminotransferase andserumtotalcholesterolareacluetopathogenesisandcandidatetargets fortreatment.HepatologyResearch:TheOfﬁcialJournalofTheJapanSociety ofHepatology2016;46:1074–87.

[82]WongVW,WongGL,ChoiPC,etal.Diseaseprogressionofnon-alcoholicfatty liverdisease:aprospectivestudywithpairedliverbiopsiesat3years.Gut 2010;59:969–74.

[83]PaisR,CharlotteF,FedchukL,etal.Asystematicreviewoffollow-upbiopsies revealsdiseaseprogressioninpatientswithnon-alcoholicfattyliver.Journal ofHepatology2013;59:550–6.

[84]McPhersonS,HardyT,HendersonE,etal.EvidenceofNAFLDprogressionfrom steatosistoﬁbrosing-steatohepatitisusingpairedbiopsies:implicationsfor prognosisandclinicalmanagement.JournalofHepatology2015;62:1148–55.

[85]ButtAA,YanP,LoRe3rdV,etal.LiverﬁbrosisprogressioninhepatitisCvirus infectionafterseroconversion.JAMAInternalMedicine2015;175:178–85.

[86]LonardoA,BellentaniS,RatziuV,etal.Insulinresistanceinnonalcoholic steatohepatitis:necessarybutnotsufﬁcient—deathofadogmafrom anal-ysisoftherapeuticstudies?ExpertReviewofGastroenterology&Hepatology 2011;5:279–89.

[87]BugianesiE,ManziniP,D’AnticoS,etal.Relativecontributionofironburden, HFEmutations,andinsulinresistancetoﬁbrosisinnonalcoholicfattyliver. Hepatology2004;39:179–87.

[88]FracanzaniAL,ValentiL,BugianesiE,etal.Riskofsevereliverdiseasein nonalcoholicfattyliverdiseasewithnormalaminotransferaselevels:arole forinsulinresistanceanddiabetes.Hepatology2008;48:792–8.

[89]LeungJC,LoongTC,WeiJL, etal. Histologicalseverityandclinical out-comesofnonalcoholicfattyliverdiseaseinnonobesepatients.Hepatology 2017;65:54–64.

[90]McPhersonS,HendersonE,BurtAD,etal.Serumimmunoglobulinlevels predictﬁbrosisinpatientswithnon-alcoholicfattyliverdisease.Journalof Hepatology2014;60:1055–62.

[91]MieleL,BealeG,PatmanG,etal. TheKruppel-likefactor6genotypeis associatedwithﬁbrosisinnonalcoholicfattyliverdisease.Gastroenterology 2008;135:282–91,e1.

[92]SingalAG,ManjunathH,YoppAC,etal.TheeffectofPNPLA3onﬁbrosis pro-gressionanddevelopmentofhepatocellularcarcinoma:ameta-analysis.The AmericanJournalofGastroenterology2014;109:325–34.

[93]SinghDK,SakhujaP,MalhotraV,etal.Independentpredictorsof steatohep-atitisandﬁbrosisinAsianIndianpatientswithnon-alcoholicsteatohepatitis. DigestiveDiseasesandSciences2008;53:1967–76.

[94]YangJD,AbdelmalekMF,PangH,etal.Genderandmenopauseimpact sever-ityofﬁbrosisamongpatientswithnonalcoholicsteatohepatitis.Hepatology 2014;59:1406–14.

[95]AnguloP,KleinerDE,Dam-LarsenS,etal.Liverﬁbrosis,butnootherhistologic features,isassociatedwithlong-termoutcomesofpatientswithnonalcoholic fattyliverdisease.Gastroenterology2015;149:389–97,e10.

[96]DongiovanniP,PettaS,MannistoV,etal.Statinuseandnon-alcoholic steato-hepatitisinatriskindividuals.JournalofHepatology2015;63:705–12.

[97]NascimbeniF,Aron-WisnewskyJ,PaisR,etal.Statins,antidiabetic medica-tionsandliverhistologyinpatientswithdiabeteswithnon-alcoholicfatty liverdisease.BMJOpenGastroenterology2016;3:e000075.

[98]vanderPoortenD,SamerCF,Ramezani-MoghadamM,etal.Hepaticfatlossin advancednonalcoholicsteatohepatitis:arealterationsinserumadiponectin thecause.Hepatology2013;57:2180–8.

[99]RoulotD,CzernichowS,LeClesiauH,etal.Liverstiffnessvaluesinapparently healthysubjects:inﬂuenceofgenderandmetabolicsyndrome.Journalof Hepatology2008;48:606–13.

[100]KoehlerEM,PlompenEP,SchoutenJN,etal.Presenceofdiabetesmellitus andsteatosisisassociatedwithliverstiffnessinageneralpopulation:the Rotterdamstudy.Hepatology2016;63:138–47.

[101]YouSC,KimKJ,KimSU,etal.Factorsassociatedwithsigniﬁcantliverﬁbrosis assessedusingtransientelastographyingeneralpopulation.WorldJournal ofGastroenterology:WJG2015;21:1158–66.

[102]ReddySK,SteelJL,ChenHW,etal.Outcomesofcurativetreatmentfor hepato-cellularcancerinnonalcoholicsteatohepatitisversushepatitisCandalcoholic liverdisease.Hepatology2012;55:1809–19.

[103]ParadisV,ZalinskiS,ChelbiE,etal.Hepatocellularcarcinomasinpatients withmetabolicsyndromeoftendevelopwithoutsigniﬁcantliverﬁbrosis:a pathologicalanalysis.Hepatology2009;49:851–9.

[104]PiscagliaF,Svegliati-BaroniG,BarchettiA,etal.Clinicalpatternsof hepatocel-lularcarcinomainnonalcoholicfattyliverdisease:amulticenterprospective study.Hepatology2016;63:827–38.

[105]GianniniEG,MarabottoE,SavarinoV,etal.Hepatocellularcarcinomain patientswithcryptogeniccirrhosis.ClinicalGastroenterologyand Hepatol-ogy:TheOfﬁcialClinicalPracticeJournaloftheAmericanGastroenterological Association2009;7:580–5.

[106]YounossiZM,OtgonsurenM,HenryL,etal.Associationofnonalcoholicfatty liverdisease(NAFLD)withhepatocellularcarcinoma(HCC)intheUnited Statesfrom2004to2009.Hepatology2015;62:1723–30.

[107]BugianesiE,LeoneN,VanniE,etal.Expandingthenaturalhistoryof nonalco-holicsteatohepatitis:fromcryptogeniccirrhosistohepatocellularcarcinoma. Gastroenterology2002;123:134–40.

[108]AschaMS,HanounehIA,LopezR,etal.Theincidenceandriskfactorsof hepatocellularcarcinomainpatientswithnonalcoholicsteatohepatitis. Hep-atology2010;51:1972–8.

[109]Sawada N,Inoue M,Iwasaki M,etal. Consumptionof n-3fatty acids and ﬁsh reduces risk of hepatocellular carcinoma. Gastroenterology 2012;142:1468–75.

[110]LonardoA,LoriaP.Potentialforstatinsinthechemopreventionand man-agement of hepatocellular carcinoma. Journal of Gastroenterology and Hepatology2012;27:1654–64.

[111] WelzelTM,GraubardBI,QuraishiS,etal.Population-attributablefractionsof riskfactorsforhepatocellularcarcinomaintheUnitedStates.TheAmerican JournalofGastroenterology2013;108:1314–21.

[112] BraviF,BosettiC,TavaniA,etal.Coffeereducesriskforhepatocellular carci-noma:anupdatedmeta-analysis.ClinicalGastroenterologyandHepatology: TheOfﬁcialClinicalPracticeJournaloftheAmericanGastroenterological Association2013;11:1413–21,e1.

[113] YangY,ZhangD,FengN,etal.Increasedintakeofvegetables,butnotfruit, reducesriskforhepatocellularcarcinoma:ameta-analysis.Gastroenterology 2014;147:1031–42.

[114] WangYG,WangP,WangB,etal.Diabetesmellitusandpoorerprognosisin hepatocellularcarcinoma:asystematicreviewandmeta-analysis.PLoSOne 2014;9:e95485.

[115] TargherG, DayCP,BonoraE.Riskofcardiovascular diseaseinpatients withnonalcoholicfattyliverdisease.TheNewEnglandJournalofMedicine 2010;363:1341–50.

[116] ByrneCD,TargherG.NAFLD:amultisystemdisease.JournalofHepatology 2015;62:S47–64.

[117] EkstedtM,HagstromH,NasrP,etal.Fibrosisstageisthestrongest predic-torfordisease-speciﬁcmortalityinNAFLDafterupto33yearsoffollow-up. Hepatology2015;61:1547–54.