This was a retrospective study carried out by the Hepatitis Study Group from the Medical Sciences School at the State University of Campinas (GEHEP). In the period 2005-2007, a total of 238 HCV non-responder patients to previous treatment with interferon plus ribavirin were considered for retreatment with PEG-IFN plus RBV. Of this total, 172 had been treated at our institution of which 152 agreed to receive retreatment with PEG-IFN plus RBV. Of these, 130 met the inclusion criteria and were retreated, in compliance with the rules created by the State of São Paulo Health Secretary and the Brazilian Ministry of Health, which provided the medication. This study was approved by our Ethical Committee. The evaluation included male and female patients, over 18, chronically infected by hepatitis C virus (HCV) genotypes 1 or 3, who were non-responders to the previous treatment with IFN-α plus RBV. Data were collected on age, sex, ethnic group, alcohol abuse and the possible infection route. Patients were considered as potentially contaminated through the parenteral route when they reported receiving transfusions, injections with non-disposable syringes or needles, or sharing materials used for manicures, acupuncture or application of tattoos given in less than optimal conditions. Patients were considered as intravenous drug users (IVDU) when reporting the use of illegal drugs or stimulants (Glucoenergan®) through the intravenous route, in groups, and sharing syringes or needles.
At baseline, all patients showed positive serum test for HCV-RNA by polymerase chain reaction (PCR-Cobas Amplicor® HCV Test - Roche Molecular Systems Inc.). All patients were negative for HBsAg and anti-HIV.
The previous treatment for HCV was carried out with IFN-α 2a or 2b (3,000,000 UI, subcutaneous, three times a week) plus RBV (administered through oral route, twice a day, weight-based dose: 1,000 mg/day for patients under 75 kg and 1,250 mg/day for those weighing over 75 kg. In Brazil this drug is provided free by the Ministry of Health in capsules of 250 mg. All patients were treated at outpatients departments. Patients with genotype 1 had been treated for over 48 weeks and those with genotype 3 were treated for over 24 weeks. Therapy was discontinued for all patients infected with genotype 1 who had positive HCV-RNA (PCR qualitative, PCR Amplicor HCV test - Roche molecular systems) at week 24. So, only non-responder patients (positive HCV-RNA at week 24) were screened for this study. Patients relapsing or who had a breakthrough in the first treatment were excluded. We included in our sample only patients who had attended every clinical evaluation of the first treatment, who had received all the drugs from the hospital pharmacy and had done all the biochemical and molecular tests requested.
Before starting the retreatment, all patients had hemoglobin > 10 g/dL, neutrophils > 1,500 cells/mm3, platelet > 70,000/mm3, albumin > 3.5 mg/dL and INR < 1.2. The bilirubin and creatinine levels were within normal values. All patients had a liver biopsy performed no more than 18 months before the start of the study, and the diagnosis was consistent with HCV. The inflammatory activity and the fibrosis grade were evaluated by the Metavir score. Patients were considered as carriers of non significant fibrosis when classified as F0, F1 and F2. Patients with F3 and F4 were considered as significant fibrosis carriers. Patients on hemodialysis, and heavy drug and alcohol users unable to comply with the treatment, were excluded. Other patients excluded had auto-immune, degenerative, renal and hematological diseases, or had other liver metabolic diseases, or those who had hypersensitivity reaction or other contraindications to the PEG-IFN plus RBV combination.
All patients were retreated with PEG-IFN α-2b associated with RBV. PEG-IFN α-2b was administered through the subcutaneous route, with a once weekly dose of 1.5 μg. All doses were administered at reference centers called "application hubs" which are outpatient departments. The 250 mg tablets of RBV were administered through the oral route, twice a day, with the dose varying according to the weight: 1,000 mg/day if the weight was less than 75 kilos and 1,250 mg/day if the weight was over 75 kilos. The retreatment period for genotype 1 was 48 weeks, while for genotype 3 it was 24 weeks.
Clinical and biochemical evaluations were performed before starting the treatment and then monthly throughout the treatment (hemogram, AST, ALT, gama-GT, TSH and free T4 dosing) in order to evaluate adverse events, tolerance and efficacy. The AST, ALT and gamma-GT levels were expressed in quotients (qALT, qAST, qGama-GT). Thus, for example, the qALT was obtained by dividing the serum ALT value by the method's highest normal value. Those with qALT > 1 had increased ALT. In the case of patients having severe adverse events or biochemical abnormalities the dose of ribavirin or PEG-IFN α-2b was reduced. In patients with hemoglobin levels lower than 8.5 mg/dL, ribavirin was suspended and when the hemoglobin levels varied between 8.5 and 10 mg/dL, the dose was reduced to half of the initial. The reduction of PEG-IFN α-2b to two-thirds of the initial dose occurred when the platelet count was lower than 30,000/mm3 or when the granulocyte count was lower than 750 cells/mm3.
The HCV-RNA was detected by PCR at weeks 12, 24 and 48 for patients infected by genotype 1 and at weeks 12 and 24 for patients infected by genotype 3. The EVR was tested at week 12 by qualitative HCV-RNA (PCR- Cobas Amplicor® HCV Test version 2.0-Roche Molecular Systems Inc.) in 104 patients, 78 (75%) with genotype 1 and 26 (25%) with genotype 3. Patients with negative PCR on this occasion were considered as complete early virologic responders (EVR). The end-of-treatment virologic response (ETVR) was tested at week 24 (genotype 3) or week 48 (genotype 1) by the same qualitative HCV-RNA test. The SVR was evaluated by a HCV-RNA test at 24 weeks after the end of the treatment.
The patients' descriptive data analysis was presented in tables for categorical variables. In order to identify risk factors for the treatment responses, uni - and multivariate Cox regression analyses were used. The adopted significance level was 5%. The computer program SAS, System for Windows (Statistical Analysis System), and version 9.1.3 Service Pack 3, SAS Institute Inc, 2002-2003, Cary, NC, USA was used.