| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Editorials |
1 Department of Laboratory Medicine, University of Washington Medical Center, Seattle, WA
2 Program in Infectious Diseases, Fred Hutchinson Cancer Research Center, Seattle, WA
aAddress correspondence to this author at: Fred Hutchinson Cancer Research Center, 1100 Fairview Avenue N, D3-100, Seattle, WA 98109. Fax 206-667-4411; e-mail kjerome{at}fhcrc.org.
Hepatits C virus (HCV) infection is a growing health problem worldwide, with more than 170 million individuals currently infected(1). Because no vaccine is currently available, the mainstay of control is treatment of infected individuals. Although the first attempts at treatment with interferon-
produced sustained responses in only 
25% of patients, more recent combination treatment regimens consisting of pegylated interferon- plus ribavirin have led to sustained response rates approaching 60%(2)(3). In spite of intense study, the specific HCV gene(s) controlling the response to combination therapy have not been identified. Instead, the best predictor is the HCV genotype of the strain present in the patient(4). HCV is an extremely diverse virus, with 6 major genotypes and more than 60 subtypes identified(5)(6). Genotype 1 is the most common genotype in the United States and Europe, and genotypes 2 and 3 are also common in these areas. Frequently, the virus within a given infected individual diverges during the course of chronic infection into multiple viral lineages with related sequences (known as quasispecies). Less frequently, an individual may become infected with a mixture of 2 or more genotypes. In spite of the wide sequence variation present in HCV, multiple studies have clearly shown a lower response rate to combination therapy with pegylated interferon- plus ribavirin for patients who are infected with genotype 1 compared with the other genotypes(7). Thus, 2 therapeutic schemes are used: a 48-week course of therapy for patients infected with genotype-1 HCV and a 24-week course for those infected with other genotypes of the virus(8). The dramatically different response rates place a premium on the accurate assignment of the HCV genotype(s) causing a given patient’s infection.
A variety of methods are available to genotype HCV-positive samples. The current gold standard method is direct sequencing of the nonstructural 5, envelope 1, or core (NS5, E1, or C) region. In contrast to these 3 regions, which contain multiple genotype-specific base pair changes over several hundred base pairs of sequence, the 5' untranslated region (5'UTR) has conserved domains and has fewer genotype-specific base pair changes(9). In spite of the lower variability in this region, at present the most commonly used methods in clinical laboratories are based on analysis of the 5'UTR by use of hybridization approaches (such as the line probe assay or melting analysis), direct sequencing of the 5'UTR, or restriction fragment length polymorphism (RFLP) analysis. The 5'UTR-based line probe assay and RFLP methods have been shown to have fair to very good agreement with sequencing of the 5'UTR, ranging from 84% to 99% for genotype and from 68% to 86% for subtype(10). However, even direct sequencing of the 5'UTR is less informative than sequencing of the NS5, E1, or C region, especially at the subtype level(11), because of the limited variability of the 5'UTR. Nevertheless, the presence of conserved domains within the 5'UTR makes PCR amplification of this region more robust than amplification of the NS5, E1, or C region; thus, the 5'UTR-based methods are more sensitive and are favored by most clinical laboratories.
In this issue, Kim et al.(12) report a promising new approach to determining HCV genotype, which they term restriction fragment mass polymorphism (RFMP) analysis. The technique is based on PCR amplification of the 5'UTR of HCV, using primers that introduce recognition sites for type IIS restriction enzymes. Type IIS restriction enzymes have the important characteristic of cleaving DNA strands at points outside their recognition site. Cleavage of the HCV PCR product by such enzymes yields multiple oligonucleotide fragments of defined length representing hypervariable regions of HCV. Kim et al. then use an elegant matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) approach to identify these hypervariable regions, allowing assignment of HCV genotype.
This novel approach promises several advantages over the current methodologies for HCV genotyping in widespread clinical use. The first is the reported 100% concordance with sequencing of the 5'UTR. Although this rate of concordance is unlikely to hold true for larger studies, and particularly for comparisons with sequencing of the NS5, E1, or C region, it compares favorably with most other methods. Perhaps more provocative is the ability of the RFMP method to detect mixed infections. The authors suggest that this method can detect minor genotypes at concentrations as low as 0.5% of the total HCV burden. Accordingly, the authors report a prevalence of mixed infections (22%) well in excess of the prevalence generally assumed. If mixed infections are truly this frequent, the ability to detect minor populations of treatment-resistant genotypes may have profound implications for HCV management.
However, several aspects of the RFMP method will require further study before this approach is widely adopted for clinical testing. Perhaps the most concerning is the reliance on analysis of the 5'UTR region. Although the RFMP method analyzes 3 different locations within the 5'UTR region, these still represent a relatively small portion of the viral genome. As noted above, analysis of the 5'UTR region rather than the NS5, E1, or C region generally leads to less accurate assignment of viral genotype and subtype, even by direct sequencing. In addition, methods that rely on limited polymorphisms present in a small region of the genome (such as the oligonucleotide fragments analyzed by MALDI-TOF MS) tend to be less informative than methods that analyze many base pair polymorphisms over a larger region of the genome (such as direct sequencing). Thus, the RFMP method may be improved by adding analysis of other regions of the viral genome, especially the NS5, E1, and C regions. Larger sample sizes will be required to determine the precise rate of agreement of the RFMP method with sequencing, particularly for samples with less common genotypes such as 4, 5, and 6.
Second, as noted above, the authors reported a very high rate (22%) of mixed-genotype samples(12). Apparently most (two-thirds) of these were confirmed by direct sequencing, and the remaining one-third were confirmed by clonal sequencing. The high rate of mixtures probably supports the authors’ assertion that the RFMP approach is more sensitive for the detection of mixed genotypes, but the rate of mixed infections seems unusually high. The RFMP approach needs to be rigorously compared with other sensitive methods for detecting mixed infections, such as clonal sequencing or heteroduplex mobility analysis, in other patient populations. Furthermore, it will be critical to establish criteria for setting the detection threshold for MS peaks that will accurately distinguish instrument noise from the signal of a minor HCV population.
Third, the authors were unable to determine the genotype of 38 of their 318 samples (12%) because they gave MALDI-TOF MS spectra that did not match any of the predicted mass patterns(12). Analysis of additional samples by RFMP and direct sequencing may identify additional patterns and thus reduce the number of nontyped samples. However, if a large percentage of specimens still cannot be assigned a genotype, clinical laboratories using the RFMP method will need to have a back-up assay for the samples.
Finally, although the authors discuss automation of their method, the method as described uses a manual spin column for isolation of RNA, 2 cycles of nested PCR (reverse transcription-PCR/PCR), and restriction enzyme digestion(12). Thus, as currently embodied, the method requires a fair amount of manual sample handling. However, all of these steps should be amenable to true automation, and if the other caveats prove not to be insurmountable, it is likely that this assay could be performed economically.
In summary, Kim et al.(12) have developed an intriguing new method (RFMP) for genotyping HCV specimens, based on MALDI-TOF MS analysis of PCR product from the 5'UTR. This is a novel approach to the problem of HCV genotyping and promises several potential advantages over current methods, most notably a greatly improved sensitivity for infections with mixed genotypes. Additional studies will be needed to conclusively demonstrate the superiority of this method. However, if the RFMP approach lives up to its promise, the ability to detect minor populations of treatment-resistant genotypes and tailor therapy appropriately may revolutionize the management of HCV infections.
References
-2b plus ribavirin: a review of its use in the management of chronic hepatitis C. Drugs 2002;62:507-556.[CrossRef][ISI][Medline]
[Order article via Infotrieve]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
