Non-invasive evaluation of liver fibrosis: current clinical use and next perspectives (chronic hepatitis C)

Transcription

1 Non-invasive evaluation of liver fibrosis: current clinical use and next perspectives (chronic hepatitis C) Paul Calès Liver and Gastroenterology department, University hospital & HIFIH laboratory, Angers University, France 21 March 2011

2 Content Introduction Current clinical use: diagnosis Available tests Meaning of tests Reproducibility Diagnostic performance Cirrhosis diagnosis Test combination Perspectives: other applications Population screening Other lesions Time related events Progression Prognosis Follow-up

3 Conflict of interest: FibroMeters family Main tests: Fibrosis Metavir Score for the 3 main causes: Virus NAFLD Alcohol CirrhoMeter (virus) Companion tests: Area of fibrosis (%) InflaMeter Alternative FibroMeters (expert system)

4 Introduction Main evaluated topic: liver fibrosis in HCV hepatitis First blood test for significant fibrosis (FibroMeter): 1997 First commercialized blood test (Fibrotest): 2002 First commercialized device (Fibroscan): : First commercialized test for cirrhosis (CirrhoMeter) Clinical availability of test combination

5 Introduction Current clinical use: diagnosis Available tests Meaning of tests Reproducibility Diagnostic performance Cirrhosis diagnosis Test combination Perspectives: other applications Population screening Other lesions Time related events Progression Prognosis Follow-up Content

6 Classification of fibrosis tests Physical methods Blood tests «Imaging» Elastometry (US, MNR) Scoring Quantitation

7 Meaning of tests Two main types of tests according to design: Constructed for a binary (yes/no) diagnostic target, usually significant fibrosis by blood test: They are indirect tests They provide a fibrosis score Directly measuring a liver signal, usually elastometry: They are direct tests They usually provide a velocimetry (transformed in elasticity reflecting liver «stiffness») Elastometry is influenced by fibrosis, activity and steatosis

8 Fibrosis score The score from 0 to 1 indicates the probability of significant fibrosis (fibrosis stage Metavir 2). For example, a FibroMeter at 0.16 means that the patient has roughly a 16% probability of having significant fibrosis.

9 Liver stiffness (kpa) Fibroscan Logarithmic scale from 2,5 to 75 kpa with Metavir stages but linear scale with fibrosis quantity: P< F N = All Metavir fibrosis stage 721 hepatitis C Area of fibrosis (%)

10 Content Introduction Current clinical use: diagnosis Available tests Meaning of tests Reproducibility Diagnostic performance Cirrhosis diagnosis Test combination Perspectives: other applications Population screening Other lesions Time related events Progression Prognosis Follow-up

11 Kappa index Calès, GCB 2008 Reproducibility according to tests The reference is liver biopsy read by expert or consensus Methods are compared for Metavir stage classification 1 0,8 0,6 0,4 0,2 0 Liver biopsy by local pathologist FibroMeter PBH Fibroscan FibroMètre

12 Content Introduction Current clinical use: diagnosis Available tests Meaning of tests Reproducibility Diagnostic performance Cirrhosis diagnosis Test combination Perspectives: other applications Population screening Other lesions Time related events Progression Prognosis Follow-up

13 Diagnostic performance Binary diagnosis (yes / no) Significant fibrosis Cirrhosis Fibrosis stage classification (several stages)

14 AUROC Performance according to 3 binary diagnostic targets 0,95 0,9 p<10-3 vs other tests Non significant differences vs other tests 0,85 0,8 0,75 FibroMeter Fibrotest Fib-4 Hepascore APRI Significant fibrosis Severe fibrosis Cirrhosis Calès et al. Liv Inter 2008, 1056 hepatitis C

15 Performance for significant fibrosis Fibrostar study, Zarski et al, EASL 2010 Test AUROC* p vs FibroMeter FibroMeter Fibrotest 0.89 NS Hepascore 0.86 <0.05 Zeng 0.85 <0.05 FIB <0.001 APRI 0,83 <0.001 ELFG 0.81 <0.001 Fibroscan 0.80 <0.05 Hyaluronate 0.78 <0.001 MP <0.001 * Weighted according to Obuchowski et al, Stat Med 2005

16 Predictibility

17 Patient rate with exclusion of cirrhosis (100% negative predictive value) 50 47, ,9 24, ,5 5,4 FibroMeter Fibrotest Hepascore Fib-4 APRI p < 10-3 between each test except for APRI vs Fib-4: p=0.13 Boursier, EJGH 2008, 1056 hepatitis C

18 Patient rate with affirmation of cirrhosis (positive predictive value 95% ) p<10-4 CirrhoMeter Calès, GCB 2008 & 2009, 1056 hepatitis C

19 Classification in fibrosis stages Fibrotest FibroMeter F0 F1 F2 F3 F4 Metavir (liver biopsy) Cut-offs F0/1 F2/3 F4 Fibroscan

20 Well classified patients (%) Calès, EASL 2010, 205 C hepatitis, reference liver biopsy: 2 experts Accuracy of fibrosis stages classification: Blood test (FibroMeter) vs liver biopsy P= ,2 64,4 P= Pathologiste expert Pathologiste 1ere ligne FibroMètre 3G Expert pathologist Local pathologist FibroMeter 3G

21 Well classified patients (%) Well classified (%) Accuracy of fibrosis stage classifications: Blood tests vs Fibroscan P < P < G FibroMeter Fibroscan Fibrotest Boursier, EASL 2010, 729 hepatitis C

22 Well classified patients (%) Bien classés (%) Accuracy of fibrosis stages classification: FibroMeter 3G vs other blood tests p < Fibrotest FibroMeter 2G FibroMeter 3G Calès et al, Clin Biochem 2010, 1056 hepatitis C

23 Content Introduction Current clinical use: diagnosis Available tests Meaning of tests Reproducibility Diagnostic performance Cirrhosis diagnosis Test combination Perspectives: other applications Population screening Other lesions Time related events Progression Prognosis Follow-up

24 Cirrhosis diagnosis The most important diagnosis For complication screening Regarding new effective treatments (DAA in HCV and NUC in HBV hepatitis) Two diagnosis Binary: cirrhosis yes or no Classification in fibrosis stages

25 CirrhoMeter Sole commercialized (blood) test for cirrhosis Useful in severe fibrosis (F3/F4) and in the follow-up of patients

26 Classification in equivalent Metavir fibrosis stages by FibroMeter 3G and CirrhoMeter 3G in chronic viral hepatitis Score of fibrosis: FibroMeter Accuracy: 86.9% Score of cirrhosis: CirrhoMeter Accuracy: 87.3%

27 Histological cirrhosis (%) Positive predictive value Distribution of histological F4 as a function of test classification (sensitivity evaluation ) 100 Fibrotest FibroMeter2G CirrhoMeter2G F0 F1/2 F1/2/3 F2 F2/3 F3 F2/3/4 F3/4 F4 All F4 Fibrosis stages by tests Calès, EASL 2011, 1056 hepatitis C

28 Patients staged F4 by test (%) Sensitivity Distribution of F4 diagnosed by test classification as a function of Metavir F stage (specificity evaluation ) 100 Fibrotest FibroMeter2G CirrhoMeter2G F1 F2 F3 F4 Metavir fibrosis stages (liver biopsy) Calès, EASL 2011, 1056 hepatitis C

29 Patients well classified (%) Binary diagnosis of cirrhosis: CirrhoMeter 2G vs Fibroscan 100 p=0.281 CirrhoMeter2G Fibroscan p= p= Accuracy Sensitivity Specificity All F4 F0-3 Metavir fibrosis stages (liver biopsy) a posteriori accuracy cut-offs Calès, EASL 2011, 679 hepatitis C

30 Content Introduction Current clinical use: diagnosis Available tests Meaning of tests Reproducibility Diagnostic performance Cirrhosis diagnosis Test combination Perspectives: other applications Population screening Other lesions Time related events Progression Prognosis Follow-up

31 Classification combining blood test and Fibroscan 2009 The combination of Fibroscan and FibroMeter provides the highest diagnostic performance in chronic liver diseases from various causes

32 Fibroscan FibroMeter Fibroscan Result (kpa) FibroMeter result (0-1) Computerization Fibroscan classification Diagnostic accuracy: 58.7% FibroMeter classification Diagnostic accuracy: 68.7% F0/1 F1/2 F2±1 F2/3 F3±1 F4 FibroMeterScan classification Diagnostic accuracy: 86.7%

33 Rate of patients (%) Comparison of FibroMeter + Fibroscan classification vs other algorithms (729 hepatitis C with liver biopsy + blood tests + Fibroscan, Boursier AJG 2011) The multiple FibroMeter + Fibroscan combination provides a high accuracy, a better precision without any liver biopsy required compared to classical algorithms

34 Introduction Current clinical use: diagnosis Available tests Meaning of tests Reproducibility Diagnostic performance Cirrhosis diagnosis Test combination Perspectives: other applications Population screening Other lesions Time related events Progression Prognosis Follow-up Content

35 Fibroscan Transient elastography as a screening tool for liver fibrosis and cirrhosis in a communitybased population aged over 45 years. Roulot Gut 2011 Conclusion: Fibroscan proved to be a useful and specific procedure to screen for cirrhosis in the general population and to detect undiagnosed chronic liver disease in apparently healthy subjects.

36 Metavir stages in patients with F3/F4 diagnosed by FibroMeter Metavir F by biopsy General Population General practitioner FibroMeter may have a 82% predictive positive value for severe liver fibrosis

37 Content Introduction Current clinical use: diagnosis Available tests Meaning of tests Reproducibility Diagnostic performance Cirrhosis diagnosis Test combination Perspectives: other applications Population screening Other lesions Time related events Progression Prognosis Follow-up

38 Metavir activity Accuracy for significant activity by InflaMeter: Well classified: 73.7% Accuracy of activity grade classification: 89.1 %

39 Introduction Current clinical use: diagnosis Available tests Meaning of tests Reproducibility Diagnostic performance Cirrhosis diagnosis Test combination Perspectives: other applications Population screening Other lesions Time related events Progression Prognosis Follow-up Content

40 Time related events Follow-up Dynamic sensitivity and reproducibility Progression rate Treatment effect Prognosis

41 Liver stiffness (kpa) Calès, Clin Biochem 2010 Dynamic sensitivity of tests in natural history of hepatitis C FibroMeter Fibroscan 1,0 12,8 10,6,4 8 6 Year 3 vs year 1: p=0.410,2 4 0,0 Year 1 Year 2 Year 3 2 Year 1 Year 2 Year 3 FibroMeter Fibroscan

42 Liver stiffness (kpa) at year 2 or 3 Dynamic reproducibility of tests in natural history of hepatitis C FibroMeter Fibroscan 1,0 16, ,6 10,4 8,2 0,0 0,0,2,4 Year r ic =0.83 r ic = ,6,8 1, Year 3 2 FibroMeter at year 1 Liver stiffness (kpa) at year 1 Calès, Clin Biochem 2010

43 Relative change (mean ± SD) Comparaison of liver morphometry and blood tests to assess fibrosis progression 101 pts, follow-up: 96 w Among histological characteristics, the relative change (%) in area of fibrosis was significantly higher than other measures. CirrhoMeter was the only blood test with a change significantly higher than that of area of fibrosis (p=0.039) Histology Blood tests * * Fibrotest Hepascore Area of fibrosis Fractal dimension Metavir F stage * AreaMeter FibroMeter CirrhoMet Calès, EASL 2011 * : significant change

44 FibroMeter Effect of treatment Progression of biological non-invasive fibrosis scores between baseline and week 72 according to the virological response. Halfon P, et al. Antivir Ther SVR: sustained virological response

45 Prognosis Naveau et al. Diagnostic and Prognostic Values of Noninvasive Biomarkers of Fibrosis in Patients with Alcoholic Liver Disease. HEPATOLOGY 2009

46 Conclusions Non-invasive tests of liver fibrosis are now included in recent guidelines for diagnosis in HCV hepatitis as first line option. Other causes are being evaluated but tests are more difficult to develop. Promising perspectives include fibrosis progression, prognosis and follow-up of patients.