Liver gene expression signature to predict response to pegylated interferon plus ribavirin combination therapy in patients with chronic hepatitis C
- T Asselah1,2,
- I Bieche2,3,
- S Narguet2,
- A Sabbagh2,3,
- I Laurendeau2,
- M-P Ripault1,2,3,
- N Boyer1,
- M Martinot-Peignoux1,
- D Valla1,
- M Vidaud2,3,
- P Marcellin1
- 1Service d’Hépatologie, Pôle des Maladies de l’Appareil Digestif and INSERM U773, CRB3, University of Paris VII, AP-HP Beaujon Hospital, Clichy, France
- 2INSERM UMR745, University of Paris Descartes, Paris, France
- 3Service de Biochimie et Génétique moléculaire, Beaujon Hospital, Clichy, France
- Dr T Asselah, Service d’Hépatologie and INSERM CRB3, Beaujon Hospital, 92 110 Clichy, France;
- Revised 3 August 2007
- Accepted 28 August 2007
- Published Online First 25 September 2007
Background and Aims: The gold standard treatment of chronic hepatitis C (CHC) is combined pegylated interferon and ribavirin. Considering side effects and treatment cost, prediction of treatment response before therapy is important. The aim of this study was to identify a liver gene signature to predict sustained virological response in patients with CHC.
Methods: Group A (training set) comprised 40 patients with CHC including 14 non-responders (NRs) and 26 sustained virological responders (SVRs). Group B (validation set) comprised 29 patients including 9 NRs and 20 SVRs. Eleven responder–relapsers were also included. A total of 58 genes associated with liver gene expression dysregulation during CHC were selected from the literature. Real-time quantitative RT-PCR assays were used to analyse the mRNA expression of these 58 selected genes in liver biopsy specimens taken from the patients before treatment.
Results: From the Group A data, three genes whose expression was significantly increased in NRs compared with SVRs were identified: IFI-6-16/G1P3, IFI27 and ISG15/G1P2. These three genes also showed significant differences in their expression profiles between NRs and SVRs in the independent sample (Group B). Supervised class prediction analysis identified a two-gene (IFI27 and CXCL9) signature, which accurately predicted treatment response in 79.3% (23/29) of patients from the validation set (Group B), with a predictive accuracy of 100% (9/9) and of 70% (14/20) in NRs and SVRs, respectively. The expression profiles of responder–relapsers did not differ significantly from those of NRs and SVRs, and 73% (8/11) of them were predicted as SVRs with the two-gene classifier.
Conclusion: NRs and SVRs have different liver gene expression profiles before treatment. The most notable changes occurred mainly in interferon-stimulated genes. Treatment response could be predicted with a two-gene signature (IFI27 and CXCL9).
Funding: This study was supported by the French National Agency for Aids and Viral Hepatitis (ANRS). TA was supported by the French Association for the Study of the Liver (AFEF). AS was supported by the French Fondation pour la Recherche Médicale.