Early detection of nonresponse to interferon plus ribavirin combination treatment of chronic hepatitis C F. J. Castro, 1 J. I. Esteban, 1 A. Jua ´rez, 1 S. Sauleda, 2 L. Viladomiu, 1 M. Martell, 1 F. Moreno, 3 H. Allende, 4 R. Esteban 1 and J. Guardia 1 1 Liver Unit, Department of Medicine, Hospital Universitari Vall d’Hebron; 2 Centre de Transfusio ´ I Banc de Teixits; 3 Roche Molecular Systems; and 4 Department of Pathology, Hospital Universitari Vall d’Hebron, Barcelona, Spain Received January 2001; accepted for publication December 2001 INTRODUCTION Two recent multicentre trials [1,2] have shown that the combination of interferon (IFN)-a2b plus ribavirin is more effective than IFN alone in the treatment of previously untreated (naive) patients with chronic hepatitis C. Unfor- tunately, even with combination therapy, close to 60% of such patients do not achieve a sustained response. In an attempt to select the best candidates for treatment, several baseline characteristics, such as age, gender, histology, genotype and baseline viral load, have been shown to be associated with the likelihood of response [1,2]. However, in clinical practice none of these response factors, either singly or in combination, is sufficiently accurate and their use in patient selection has not been shown to be cost-effective when compared with immediate empirical treatment [3–5]. Accordingly, the use of some baseline variables (genotype and viral load) is restricted to establishing the duration of therapy [6]. Several studies have shown a strong association between clearance of serum HCV RNA at week 4–12 of IFN mono- therapy and sustained virological response [7–13]. However, implementation of monitoring early virological response has been hampered by the lack of standardization of the tests used in most of these studies and by the advent of combination therapy. Indeed, it has been suggested that dynamics of early viral response during combination therapy is different from that observed during monotherapy, with a very high pro- portion of sustained responders to combination therapy still having detectable HCV RNA 12 and even 24 weeks after treatment onset [1]. This difference, however, is more apparent than real and is due to the combined effect of two factors: the higher (10–100-fold) sensitivity of the RT-PCR tests used during combination therapy trials as compared Abbreviations: ALT, alanine aminotransferase; HCV, hepatitis C virus; NPV, negative predictive value; NR, nonresponders; PPV, positive predictive value; ROC, receiver operating characteristic; Se, sensitivity; Sp, specificity; SR, sustained responders. Correspondence: J. I. Esteban, Hospital Universitari Vall d’Hebron, Servei de Medicina Interna-Hepatologia, Pg Vall d’Hebron, 119– 129, 08035 Barcelona, Spain. E-mail: esteban@hg.vhebron.es Journal of Viral Hepatitis, 2002, 9, 202–207 Ó 2002 Blackwell Science Ltd SUMMARY. We have investigated the value of early hepatitis C virus (HCV) RNA decline (DHCV RNA) to predict response to combination therapy in 66 chronic hepatitis C patients treated with IFN-a2b (3 MU thrice weekly) and ribavirin (800 mg daily) for 12 months [25 sustained responders (SR) and 41 nonresponders or relapsers (NR)]. Serum HCV RNA was retrospectively measured in samples obtained at baseline and 4, 8 and 12 weeks after treatment onset, using a commercially available quantitative RT-PCR assay. At 4 weeks, serum HCV RNA had decreased a mean of 2.6 ± 0.8 logs among SR as compared with only 0.5 ± 0.8 logs in NR (P < 0.001), and was already undetectable (< 600 IU/mL) in 12 (48%) of the SR but in none of the NR. At 8 weeks, HCV RNA was undetectable in 21 SR and in 2 NR and mean DHCV RNA were 4.2 ± 1.3 and 0.8 ± 1.0 logs, respectively (P < 0.001). At week 12 all SR had undetectable HCV RNA as compared with only five NR (P < 0.001). Stepwise logistic regression analysis identified DHCV RNA at 12 weeks as the strongest predictor of sustained response. Receiver operating characteristic (ROC) curves of DHCV RNA for sustained response prediction identified sensitivity peaks with 100% negative predictive value corresponding to DHCV RNA > 1 log at 4 weeks, > 2 logs at 8 weeks and > 3 logs at 12 weeks. Our results show that early changes in the HCV RNA level may reliably identify patients having no chance of a sustained virological response during the first 3 months of com- bination therapy, thus providing an excellent tool for optimizing antiviral treatment of chronic hepatitis C. Keywords: combination therapy, HCV viral load, quantitative RT-PCR, response prediction, therapy monitoring, viral dynamics.