The distribution and kinetics of hepatitis C virus INTRODUCTION (HCV) genotypes and the prevalence of mixed Hepatitis C virus (HCV) is a positive-stranded RNA infections were studied in a group of 45 French virus of 9,400 nucleotides [Choo et al., 1989] with homolpatients with haemophilia A or B or von Willeogy to the flavivirus family. The genome includes a 5Ј brand's disease, 21 of them being anti-human untranslated region (5Ј UTR) [Bukh et al., 1992], a long immunodeficiency virus (HIV) positive; genotypsequence encoding for a polyprotein processed seconding was carried out by three methods based on arily in structural (core and envelopes [E1-E2]) and the core, 5Ј untranslated region (5ЈUTR), and the non-structural (NS2-NS5) proteins and a 3Ј UTR of detection of type-specific NS4 antibodies. Genovariable length [Choo et al., 1991; Takamizawa et al., typing of the 5ЈUTR revealed genotypes 1a 1991; Matsuura and Miyamura, 1993]. The 5ЈUTR and (n ϭ 10), 1b (n ϭ 13), 2a (n ϭ 3), 2b (n ϭ 4), 2NC
core are highly and well-conserved sequences. In con-(n ϭ 3), 3a (n ϭ 10), and two mixed infections trast, the putative envelope genes revealed much more (1a ϩ 1b and 3a ϩ 2). Five of 33 patients showed significant divergence: in particular, "hypervariable" rea change from one HCV genotype to another.
gions at the N-terminal part of E2 have been demon-The core genotyping assay showed 8 of 45 mixed strated [Higashi et al., 1993; Okamoto et al., 1991]. infections: 6/8 1a ϩ 1b and 2/8 3a ϩ 2. Sequenc-
The HCV genome shows significant variability which ing of core polymerase chain reaction (PCR)
was proven by comparative analyses of the numerous products showed that mixed infection 1a ϩ 1b full-length and partial sequences of various isolates in could be explained by nonspecific annealing of distinct geographical areas. Although there is no final the 1b primer to type 1a sequence. By designing agreed classification of genotypes, it is generally acnew primers whose sequence was more specific cepted that there are six major genotypes and numerous to HCV types 1a and 1b, we could confirm 1a ϩ 1b
subtypes [Bukh et al., 1994; Simmonds et al., 1994a,b]. mixed infection in only one of six cases. Serotyp-Several studies have shown a correlation between HCV ing assay showed for 17 of 21 anti-HIV negative genotypes and the rate of response to interferon (IFN) patients a concordance with the 5ЈUTR genotype;
treatment [Yoshioka et al., 1992;Kanai et al., 1992]. In however, only 6 of 19 anti-HIV positive patients addition, there is evidence that HCV type 1b might be showed detectable serological reactivity. In sumassociated with more severe liver diseases, but this is mary, we have observed a similar HCV genotype still a debated issue [Pozzato et al., 1992; Fe ´ray et al., distribution between our haemophilic group and1992, 1993;Nousbaum et al., 1995]. the French anti-HCV positive patients. The study HCV is now the most common cause of post-transfudemonstrates the difficulties of assessing with sion and sporadic non-A/non-B hepatitis (NANBH). The the presently available genotyping and serotypintroduction of clotting factor concentrates in the 1970s ing assays the real prevalence of mixed infecis largely responsible for the transmission of HCV in haetions in multiply transfused patients.