传染病学传染病学 (9).pdf
Association between positivity of serum autoantibodies and liverdisease severity in patients with biopsy-proven NAFLDYu-Jie Zhoua,Kenneth I.Zhengb,Hong-Lei Mab,Gang Lib,Xiao-Yan Panc,Pei-Wu Zhud,Giovanni Targhere,Christopher D.Byrnef,Xiao-Dong Wangb,g,h,Yong-Ping Chenb,g,h,Xiao-Bo Lia,Ming-Hua Zhengb,g,h,*aDivision of Gastroenterology and Hepatology,Key Laboratory of Gastroenterology and Hepatology,Ministry of Health,Renji Hospital,School ofMedicine,Shanghai Jiao Tong University,Shanghai Institute of Digestive Disease,Shanghai,ChinabNAFLD Research Center,Department of Hepatology,The First Affiliated Hospital of Wenzhou Medical University,Wenzhou,ChinacDepartment of Endocrinology,The First Affiliated Hospital of Wenzhou Medical University,Wenzhou,ChinadDepartment of Laboratory Medicine,The First Affiliated Hospital of Wenzhou Medical University,Wenzhou,ChinaeSection of Endocrinology,Diabetes and Metabolism,Department of Medicine,University and Azienda Ospedaliera Universitaria Integrata of Verona,Verona,ItalyfSouthampton National Institute for Health Research Biomedical Research Centre,University Hospital Southampton,Southampton General Hospital,Southampton,UKgInstitute of Hepatology,Wenzhou Medical University,Wenzhou,ChinahKey Laboratory of Diagnosis and Treatment for the Development of Chronic Liver Disease in Zhejiang Province,Wenzhou,ChinaReceived 18 May 2020;received in revised form 3 October 2020;accepted 6 October 2020Handling Editor:A.SianiAvailable online 13 October 2020KEYWORDSAutoantibodies;Fibrosis;NAFLD;Liver biopsyAbstractBackground and aims:Some previous studies reported serum autoantibody positivityin patients with nonalcoholic fatty liver disease(NAFLD).The clinical significance of these find-ings remains uncertain.We aimed to investigate the association between the presence of serumautoantibodies and liver disease severity in NAFLD.Methods and results:A total of 388 consecutive patients with biopsy-proven NAFLD wereincluded in the study.Various serum autoantibodies(including also anti-nuclear antibodiesANA)were detected by indirect immunofluorescent or immunoblotting assays.Overall,84(21.6%)patients with biopsy-confirmed NAFLD had positivity for at least one of the measuredserum autoantibodies.ANA positivity was present in 50(12.9%)patients,whereas anti-U1RNPor pANCA antibodies were detectable in 9(2.3%)and 6(1.5%)patients,respectively.Multivariatelogistic regression analysis showed that ANA positivity(adjusted-odds ratio:4.51,95%CI:1.77e11.5;P Z 0.002)or positivity of any serum autoantibodies(adjusted-odds ratio:3.14,95%CI:1.30e7.62;P Z 0.01)were independently associated with advanced liver fibrosis(stages F3eF4).In serum autoantibody/ANA-positive patients,the proportion of those with advancedfibrosis was also greater among carriers of PNPLA3 rs738409 GG or CG than among those carryingPNPLA3 rs738409 CC genotype.Conclusions:Serum autoantibody positivity was independently associated with advanced liverfibrosisinpatientswithbiopsy-provenNAFLD.ThepresenceofserumautoantibodiesinpatientswithAbbreviations:NAFLD,nonalcoholic fatty liver disease;AIH,autoimmune hepatitis;NASH-CRN,nonalcoholic steatohepatitis clinicalresearch network;ANA,anti-nuclear antibodies;PNPLA3,patatin like phospholipase domain containing 3;PBC,primary biliary chol-angitis;anti-CENP-B,anti-centromere antibody;AHA,anti-histone antibody;AMA,anti-mitochondrial antibody;SLA,soluble liver an-tigen;ASMA,anti-smooth muscle antibody;LKM,liver-kidney microsomal antibody;LC-1,liver cytosol antibody type 1;ELISA,enzyme-linked immunosorbent assay;HOMA-IR,homoeostasis model assessment of insulin resistance;SNP,single nucleoid polymorphism;MALDI-TOF,Matrix-Assisted Laser Desorption/Ionization-Time of Flight;IFN,interferon;TNF,tumor necrosis factor;IL,interleukin.*Corresponding author.NAFLD Research Center,Department of Hepatology,The First Affiliated Hospital of Wenzhou Medical University,No.2Fuxue Lane,Wenzhou,325000,China.Fax:86 577 55578522.E-mail address:(M.-H.Zheng).https:/doi.org/10.1016/j.numecd.2020.10.0040939-4753/2020 The Italian Diabetes Society,the Italian Society for the Study of Atherosclerosis,the Italian Society of Human Nutrition and the Department of ClinicalMedicine and Surgery,Federico II University.Published by Elsevier B.V.All rights reserved.Nutrition,Metabolism&Cardiovascular Diseases(2021)31,552e560Available online at Nutrition,Metabolism&Cardiovascular Diseasesjournal homepage: fibrosis occurred more frequently amongst those carrying PNPLA3 rs738409 GG or CG ge-notypes.2020TheItalianDiabetesSociety,theItalianSocietyfortheStudyofAtherosclerosis,theItalianSo-cietyofHumanNutritionandtheDepartmentofClinicalMedicineandSurgery,FedericoIIUniversity.Published by Elsevier B.V.All rights reserved.BackgroundNonalcoholic fatty liver disease(NAFLD)is one of the mostcommon causes of chronic liver diseases worldwide,andposes a health threat in nearly a quarter of the Asian adultpopulation 1.Serum autoantibodies refer to the immu-noglobulins reactive to autoantigens inside,on the surface,or outside the cell,which reflect autoimmunity and mayserve as biomarkers for many autoimmune disorders.Although the pathogenesis of NAFLD is strongly associatedwith overweight/obesity,insulin resistance and othermetabolic risk factors rather than autoimmunity 2,3,thepresence of one or more serum autoantibodies may occurin 20e35%of patients with NAFLD,who do not havecoexisting autoimmune hepatitis(AIH)4e6.Previous reports examining the clinical significance ofserum autoantibodies in individuals with NAFLD haveyielded inconsistent results.For example,a study of 225United States patients with biopsy-proven NAFLD hasshown that serum autoantibody positivity was associatedwith greater histological severity of NAFLD(especiallyliver fibrosis and necro-inflammation)4.In contrast,inthe nonalcoholic steatohepatitis clinical research network(NASH-CRN)database,Vuppalanchi et al.did not find anyassociation between the presence of serum autoanti-bodies and more advanced histologic features of NAFLDin a cohort of 864 patients with biopsy-proven NAFLD7.Additionally,Ravi et al.have shown that serumautoantibody positivity was not associated with clinicalpresentation and clinical outcomes in patients withNAFLD or alcohol-related liver disease in the USA 6.Another small study of 84 Italian patients with NAFLDfoundthatanti-nuclearantibodies(ANA)positivity(1:100 titers)was associated with markers of insulinresistance 5.To our knowledge,in Asian populations,there is onlyone published study performed in 212 Japanese patientswith NAFLD,that did not find any association betweenseverity of NAFLD histology and serum positivity for ANA8.To date,studies examining the association betweenautoantibody positivity and NAFLD severity in the Chinesepopulation are lacking.Since heterogeneity may exist be-tween different ethnic groups,it is important to investi-gate whether there are associations between positivity ofserum autoantibodies and liver disease severity acrossdifferent ethnic groups.It is well known that the patatin like phospholipasedomain containing 3(PNPLA3)I148M genotype(rs738409)is the most robust single genetic variant influencingdisease severity and progression of NAFLD 9,10,but it iscurrently not known whether this genetic variant is alsoassociated with serum autoantibody positivity in NAFLD.Since it remains uncertain whether the presence ofserum autoantibodies is associated with liver diseaseseverity in patients with NAFLD,the main aim of our studywas to investigate the association between serum auto-antibodies and liver disease severity in a cohort of Chinesepatients with histologically-proven NAFLD.MethodsStudy populationWe consecutively recruited adult individuals with biopsy-proven NAFLD at the First Affiliated Hospital of WenzhouMedical University between December 2016 and January2019.Detailed inclusion and exclusion criteria have beendescribed previously 11,12.Briefly,from an initial cohortof 739 patients with suspected NAFLD(based on the im-aging evidence of hepatic steatosis and/or elevated serumliverenzymes),weruledoutcaseswithotherknowncausesof liver disease,such as alcohol-related liver disease,drug-induced hepatitis,viral hepatitis,autoimmune hepatitis(AIH),Wilsons disease and primary biliary cholangitis(PBC).Patients with history of significant alcohol con-sumption(20 g/day in women and 30 g/day in men)were excluded by a questionnaire-based lifestyle survey.We also excluded some patients with serum anti-Ro52 oranti-M2-3E positivity,who probably deserve more in-depth assessment of other coexisting autoimmune/rheu-matic diseases.As a consequence,a total of 388 patients with biopsy-proven NAFLD,who did not report any prior history ofautoimmune or rheumatic diseases,were included in thefinal analysis.Liver histology data confirmed there were nopathologic features of AIH(interface hepatitis,plasma cellinfiltration,hepatic rosette formation,or emperipolesis)13 or PBC(chronic non-suppurative destructive chol-angitis,florid duct lesions)14 in any of these patients.Inparticular,the AIH score proposed by the internationalautoimmune hepatitis group scoring system 12 was lessthan 6 in all patients enrolled in this study.Ethical approval was obtained from the First AffiliatedHospital of Wenzhou Medical University Ethics Committee(2016-246,approved on 1 December 2016).The studyprotocol conforms to the ethical guidelines of the 1975Declaration of Helsinki.Each subject has provided writteninformed consent.Serum autoantibodies in NAFLD553Clinical parameters and autoimmune markersmeasurementFor each patient,fasting blood samples were obtained onthe same day of the liver biopsy examination.Serum au-toantibodies,including ANA,anti-mitochondrial antibody(AMA),anti-dsDNA,anti-smooth muscle antibody(ASMA),andperinuclearanti-neutrophilcytoplasmicantibody(pANCA),were determined by indirect immunofluorescentassay.The substrate for ANA was human epithelial-2(HEp-2)cells,the substrate for ASMA testing was rat stomachtissue,and the substrate for pANCA was neutrophilicgranulocytes.Positiveimmunofluorescenttestsweredefined as presence of specific yellowish green fluores-cence in cells or tissues.ANA with a titer?1:100 wasconsidered positive,as suggested by Qiu et al.and theChinese consensus on the diagnosis and management ofAIH 15,16.Anti-U1RNP,anti-Sm antibody,anti-solubleliver antigen(anti-SLA),anti-Ro52,anti-nucleosome anti-body,anti-liver-kidney microsomal antibody(anti-LKM),anti-liver cytosol antibody type 1 antibody(anti-LC-1),anti-PML protein antibody,anti-SSA,anti-SSB,anti-Scl-70,anti-Jo-1,anti-Rib-P,anti-centromereantibody(anti-CENP-B),anti-histone antibody(AHA),AMA-M2,anti-M2-3E,anti-gp210 antibody,and anti-sp100 antibody weremeasuredbycommerciallyavailableimmunoblottingassay kits.The substrate for immunoblot tests was nitro-Blue-Tetrazolium/5-bromo-4-chloro-3-indolyl-phosphate(NBT/BCIP),and positive immunoblot tests were defined asdetection of a clearly visible mark corresponding to thespecific position on the blot.The positivity of autoimmunemarkers was defined as the presence of at least one of theaforementioned serum autoantibodies.Other routine blood biochemistry assessments wereperformed by standard laboratory methods,for which adetailed description can be found in our previous studies12.Anthropometric measurements,including body massindex(BMI,weight/height2),waist circumference and hipcircumference were measured in all patients.Hyperten-sion was diagnosed based on blood pressure?140/90 mmHg or use of any antihypertensive drugs.Homoeo-stasis model assessment(HOMA-IR)-estimated insulinresistance was calculated in all non-diabetic patients withNAFLD as follows:fasting insulin(mU/mL)*fasting glucose(mmol/L)/22.5.A diagnosis of type 2 diabetes mellitus wasbasedonmedicalhistory,fastingglucoselevels?7.0 mmol/L and/or glycated hemoglobin?6.5%17.Histological evaluation of NAFLDUltrasound-guided percutaneous liver biopsy was per-formed using 16G Hepafix needle.Formalin-fixed paraffin-embedded biopsy tissue with hematoxylin-eosin andMassons trichrome staining were interpreted by anexperienced hepato-pathologist(Xiao-Dong Wang),whowas blinded to patients clinical and laboratory data.His-tological scoring of NAFLD was based on the NASH-CRNscoring system 18.We defined NASH as presence ofsteatosis,ballooningandlobularinflammationonhistology(at least one point for each)and NAFLD activityscore(NAS)?4 18.The histological staging of liverfibrosis was determined according to the Brunts criteria19.Presence of stages F3eF4 on histology was defined asadvanced liver fibrosis,a condition which increases risk ofhepatocellular carcinoma and is associated with poorerprognosis in NAFLD 20.Genetic analysisWe extracted 20 ng genomic DNA from peripheral bloodsamples for single nucleoid polymorphisms(SNP)geno-typing of PNPLA3 rs738409.Genotyping assay for SNP lociwere designed via the MassARRAY System(Agena Biosci-ence,San Diego,CA).After amplification of DNA samplesthrough locus-specific polymerase chain reaction(PCR)according to Assay Design Suite software(Version 3.1),PNPLA3 rs738409 allele detections were performed byMatrix-Assisted Laser Desorption/Ionization-Time of Flight(MALDI-TOF)mass spectrometry.Statistical analysisContinuous variables were presented as mean?standarddeviation or median interquartile range,and categoricaldatawerepresentedasnumber(%).Continuousdatawithorwithout normal distribution were compared using the un-pairedStudentst-testorthetwo-sampleManneWhitneyUtest,as appropriate.Categorical variables were comparedusing the chi-squared test or the Fishers exact test,asappropriate.The Cochran-Armitage trend test(chi-squaredtest for trend)was employed to examine the presence ofassociation between serum autoantibody positivity andhistological scoring of NAFLD via the“CATT”R package,which has better accuracy than the Pearsons chi-squaredtest for the assessment of associations between a variablewith two categories and an ordinal variable with?3 cate-gories21,22.StatisticalanalyseswereconductedusingIBMSPSS version 22.0 and R version 3.6.0(https:/www.r-project.org/).Forallstatisticaltests,atwo-sidedPvalue?0.05 was considered statistically significant.ResultsPatient characteristicsA total of 739 consecutive patients with biopsy-provenfatty liver disease of any aetiology were initially enrolled.We excluded patients with alcoholic fatty liver(n Z 105),those with probable or definite AIH(n Z 8;according tothe international AIH group scoring system),drug-inducedhepatitis(nZ31),hepatitisvirus-infectedpatients(n Z 181),and those(n Z 26)with either serum anti-Ro52or anti-M2-3E positivity,who probably deserve more in-depth assessment for autoimmune/rheumatic diseases.Thus,a total of 388 patients with biopsy-confirmed NAFLD,who did not have any prior history of autoimmune orrheumatic diseases,were included in the final analysis.554Y.-J.Zhou et al.Table 1 shows the clinical,biochemical and liver hist
