FibroTest in the diagnosis of HBV

1
FibroTest in the diagnosis of HBV

• Publications on diagnostic performance

2
In this Presentation
1. Diagnosis and clinical options

2. First validation of FibroTest-ActiTest in HBV

3. FibroTest in histological changes

4. FibroTest, combinations and comparison with other non invasive methods

5. Meta-analysis
3
Diagnosis and clinical options
Positive serology
Poynard et al, Comp Hepatol 2004
4
First HBV validation
5
Myers RP et al, J Hepatol 2003 First
Validation in HBV

• Prediction of liver histological lesions with
  biochemical markers in patients with chronic
  hepatitis B (n209)

Conclusions
Sensibility analyse markers nt affected by ethnicity, HBV DNA or HBV status
In AgHbe positive patients FT more accurate than AST (AUROC 0,89 vs 0,79
AST vs METAVIR Inflammation grade ActiTest vs METAVIR Inflammation grade In AgHbe positive patients FT more accurate than AST (AUROC 0,89 vs 0,79
In AgHbe positive patients FT more accurate than AST (AUROC 0,89 vs 0,79
FibroTest useful for identification of HBV-related fibrosis
ActiTest useful for excluding significant necroinflammation
AST vs METAVIR Fibrosis stage ActiTest vs METAVIR Fibrosis stage ActiTest useful for excluding significant necroinflammation
6
FibroTest in Histological changes
7
Poynard et al, Am J of hepatology 2005

• Longitudinal Assessment of Histology Surrogate
  Markers (FibroTestActiTest) During Lamivudine
  Therapy in Patients with Chronic Hepatitis B
  Infection

Conclusion
In patients with chronic hepatitis B, a 24-month course of lamivudine treatment leads to a significant decrease in necroinflammatory grades and fibrosis stages as assessed by noninvasive markers, with the occurrence of a three-phase kinetics.
FTAT should be useful in the noninvasive follow-up of lamivudine treatment.
AUROC of FirboTest ActiTest 0,74-077, similar as the one observed in patients with HCV
8
Poynard et al Am J G 2005

• Kinetics of fibrosis according to baseline stages
  in HBV patients
• treated with lamivudine 2 years (n283)

Conclusion
44 Cirrhosis 42 (95) improvement at 24 months
Significant regression (gt0.30) in 14/44 (32)
F2F3F4 P0.01
F0F1 NS
9
Poynard et al, AASLD 2007
Impact of adefovir dipivoxil on liver fibrosis
and activity assessed with biochemical markers
(FibroTest-ActiTest) in patients infected by
Hepatitis B Virus
Study group Chronic hepatitis B (HBeAg and HBeAg-) Randomized in two placebo-controlled trials of ADV Available paired liver biopsies and FibroTest-ActiTest at baseline and after 48 weeks of treatment Liver biopsies scored for fibrosis and inflammation, utilizing Knodell, Ishak and METAVIR scoring systems, one blinded central pathologist
Methods AUROCs for the diagnosis of advanced fibrosis, cirrhosis, and moderate-severe activity Sensitivity analyses ethnicity, biopsy size, HBeAg status Impact of treatment assessed on liver injury (biopsy and FibroTest-ActiTest) according to baseline stage, and virological response Analysis of discordance between biopsy and FibroTest
Poynard et al, AASLD 2007
10
Poynard et al, AASLD 2007 - Results

• FibroTest and Fibrosis Stages

Ishak Stages METAVIR stage
11
Poynard et al, AASLD 2007 - Results

• ActiTest and Necro-Inflammatory Features

Peri Portal NecrosisKnodell Score Lobular necrosis Knodell Score Portal InflammationKnodell Score
12
Poynard et al, AASLD 2007 - Results

• ActiTest and Necro-Inflammatory Scoring System

Ishak Activity grade METAVIR Activity grade
13
Poynard et al, AASLD 2007 - Results

• Impact of HBV treatment on fibrosis Biopsy
  versus FibroTest
• 48 weeks with adefovir (n304) or placebo (n158)

Biopsy
FibroTest
Plt0.0001
14
Poynard et al, AASLD 2007 - Results

• Impact of HBV treatment on fibrosis in HBV
  Virological Responders with advanced baseline
  fibrosis
• 97 treated with adefovir, 9 treated with placebo
  (spontaneous clearance)

Biopsy
FibroTest
Plt0.0001
15
Poynard et al, AASLD 2007 - Conclusions
Discordance analysis 29 discordances estimated by the classical analysis considering biopsy as the gold standard 29 discordant cases had incoherence between virological response and histological response Failure attributable to biopsy 66 (19/29) false positive median 11mm, false negative median 7-mm Failure attributable to FT-AT 34 (10/29) If these estimates are true the real rates of patients misclassified using FT-AT is 10 (34 of 29)
Conclusions Provides an accurate quantitative estimate of liver fibrosis and necro-inflammatory activity Is effective and very sensitive as noninvasive marker of histological changes during treatment or followup without treatment
16
Combination and comparison with other non
invasive methods

• FibroScan, APRI, Mp3

17
Sebastiani et al, J Hepatol 2006

• Diagnostic performance of non-invasive biomarkers
  of liver fibrosis in chronic hepatitis B (n110)

Results
Diagnosis of F2F3F4 F4 Se 89,5 62,5 Sp 78,8 98,4 NPV 64,7 95,4 PPV (for F4) 83 FibroTest correctly classfied all patients
Conclusions
Fibrotest presents with the best accuracy in all the subgroups of patients with chronic liver disease Combination of markers should reduce the need for liver biopsy
HEPATITIS B (AUC) HEPATITIS B (AUC)
APRI FIBROTEST
0.72 0.85
0.64 0.76
gtF2
F4
18
Sebastiani et al, J Hepatol 2006 Safe Biopsy

• Sequential Algorithms for Fibrosis Evaluation
  (SAFE BIOPSY) Stepwise modelling aimed to achive
  accuracygt 95

For significant fibrosis For cirrhosis
19
Sebastiani et al, J Hepatol 2006 Safe Biopsy

• Sequential Algorithms for Fibrosis Evaluation
  (SAFE BIOPSY) INTERIM ANALYSIS ON 210 HBV CASES

SAFE BIOPSY for SIGNIFICANT FIBROSIS SAFE BIOPSY for CIRRHOSIS
Accuracy () 96 90
Saved biopsies () 45 77
Saved cost () 44 75
20
Castera L. et al, J Hepatol 2006

• Prospective comparison in FibroScan (FS) and
  FibroTest (FT) in inactive hepatitis B carriers

Study Group Cohort of 154 HBV patients, among these 40 inactive carriers
Method FibroTest and FibroScan given the same day
Results Fibroscan Failure 6 Median value (FS and FT) significantly lower in inactive carriers than in other patients Agreement of FS and FT for the absence of significant fibrosis in 83 of the patients
Conclusion Non invasive assessment of fibrosis in HBV inactive carriers per FT and FS could be useful
21
Hilleret et al, J Hepatol 2006

• Diagnostic accuracy of mp3 score compared to
  hyaluronate and FibroTest for evaluating liver
  fibrosis in chronic hepatitis B

Diagnostic accuracy evaluated by AUROC for discriminating F0F1F2 vs F3F4 Diagnostic accuracy evaluated by AUROC for discriminating F0F1F2 vs F3F4 Diagnostic accuracy evaluated by AUROC for discriminating F0F1F2 vs F3F4 Diagnostic accuracy evaluated by AUROC for discriminating F0F1F2 vs F3F4
HA MP3 FT Comments MP3 score greater than 0.50 had a PPV for extensive fibrosis of 82, while score lower than 0.30 had a NPV of 88. When combining MP3 (0.40) and HA (80), the PPV increased to 92 for F3F4
0.82 0.81 0.81 Comments MP3 score greater than 0.50 had a PPV for extensive fibrosis of 82, while score lower than 0.30 had a NPV of 88. When combining MP3 (0.40) and HA (80), the PPV increased to 92 for F3F4

Conclusions Conclusions Conclusions Conclusions
MP3, HA and FT have a good accuracy in HBV infection in predicting extensive fibrosis, especially when used in combination. Especially useful for of inactive carriers who might have cirrhosis. MP3, HA and FT have a good accuracy in HBV infection in predicting extensive fibrosis, especially when used in combination. Especially useful for of inactive carriers who might have cirrhosis. MP3, HA and FT have a good accuracy in HBV infection in predicting extensive fibrosis, especially when used in combination. Especially useful for of inactive carriers who might have cirrhosis. MP3, HA and FT have a good accuracy in HBV infection in predicting extensive fibrosis, especially when used in combination. Especially useful for of inactive carriers who might have cirrhosis.
22
Meta analysis
23
Poynard et al, clin chem 2007
FibroTest Meta-Analysis 30 Published
Studies 6.378 Patients 2001-2006 AUROC0.84
(0.83-0.86) for F2F3F4
The best you can obtain with 20mm biopsy is 0.90
Bedossa 2003