Antibiotics in Laboratory Medicine, 6 Ed.

Chapter 11. Antiviral Agents for HIV, Hepatitis, Cytomegalovirus, and Influenza: Susceptibility Testing Methods, Modes of Action, and Resistance

Daniel Amsterdam

Antiviral compounds have now been available for several decades. However, more recently, there has been a surge in the development of antivirals sparked by the AIDS epidemic and influenza pandemics. As was the case for antimicrobial agents, the question of their evaluation in clinical practice is an ongoing challenge. The correlation of in vitro activity, determined either by phenotypic or genotypic assays with patient outcome required continued research and outcome determinations. Although the initial approach for determining in vitro activity was to identify a phenotype, antiviral evaluations were conducted using genotypic assays. This was made possible through sequencing techniques that were applied to portions of viral RNA and DNA related to the targets of available compounds. As will be seen, phenotypic testing is a more complex, time-consuming, and costly procedure.

The representative viruses reviewed in this chapter establish chronic infection in their hosts. They invade and, as such, behave as quasispecies—that is, complex mixtures of genetically related but distinct viral populations in equilibrium within a given replicative environment. For this reason, at any given time, patients harbor a large number of different viral genomes, which can lead to unreliable evaluations of susceptibility/resistance when using phenotypic methodologies.

Over the past decade, the field of molecular virology testing has evolved to provide broadened dynamic range and increased assay sensitivity, leading to more actionable decision points for clinicians. Viral load monitoring of HIV, the hepatitis viruses, and human cytomegalovirus, is a routine barometer of a patient’s infected viral status and is typically used by physicians to gauge a patient’s response to therapy.

Unlike the case for bacterial pathogens where antimicrobial susceptibility testing is typically routine, antiviral susceptibility testing, because of complexity and cost, has more limited applicability. Antiviral susceptibility testing is necessary for evaluating new antiviral agents, defining mechanisms of resistance, determining the frequency of emergence of drug-resistant viral mutants, and for testing cross-resistance to alternative agents.

Aside from spontaneous mutations, most of the available agents are able to induce mutations, at least at their active site, resulting in modified viral replication capacity. This varies according to mutations, some of which may be beneficial in reducing viral replication (i.e., viral fitness). However, reciprocal interaction between mutations can only be measured by phenotypic testing. For the viral infections presented in this chapter (i.e., HIV, hepatitis B and C, CMV, and influenza), phenotypic and genotypic approaches will be presented.

Currently, there are limited standards for performing antiviral susceptibility testing. The Clinical and Laboratory Standards Institute (CLSI) has published one approved standard applicable to herpes simplex virus (HSV) (1), which will not be addressed in this chapter.

Variables that influence the final results of these assays include cell culture line, viral inoculum titer, incubation temperature and interval, concentration range of the antiviral agent studied, reference strains, assay method, and end point determination and interpretation. The virus inoculum titer has been shown to affect the end point evaluation; a large inoculum can render a susceptible isolate appear resistant, whereas a reduced inoculum can make all isolates appear susceptible (2). The duration of incubation must be sufficient to permit detection of small plaques in the plaque reduction assay (PRA) or to allow growth of a potentially resistant subpopulation, which may replicate at reduced rates compared to the wild-type virus. The important consideration of the viral inoculum is the heterogeneity within the virus population. A single clinical isolate can represent a mixture of drug-susceptible and drug-resistant phenotypes (3–5).

Phenotypic versus Genotypic Assays

Phenotypic assays evaluate the inhibitory effect of antiviral agents on the virus population recovered from a patient. Various end point determinations have been measured and include plaque reduction as in the PRA; inhibition of viral nucleic acid synthesis; reduction in the yield of viral structural proteins such as the p29 antigen of HIV or the hemagglutinin (HA) of influenza; and the diminution in the enzymatic activity of a functional protein, for example, the reverse transcriptase (RT) of HIV-1 and the neuraminidase (NA) of influenza virus. Currently used phenotypic assays include PRA, dye uptake (DU), DNA hybridization, enzyme immunoassay (EIA), NA inhibition and yield reduction assays (YRAs), and peripheral blood mononuclear cell (PBMC)–based cocultivation and recombinant virus assay (RVA) for HIV-1. It should be noted that all assays have not been used for every viral group. Genotypic assays evaluate viral nucleic acid to detect the presence of specific sequence that causes antiviral drug resistance. Genotyping has been applied to all virus groups discussed in this chapter. The approach for genotypic assays encompass nucleic acid sequencing by automated sequencers, PCR amplification and restriction enzyme digestion of the products, and hybridization of microarrays of oligonucleotide probes. Phenotypic assays assess the combined effect of multiple resistant mutations (if present), whereas genotypic assays can only detect resistance mutation within the selected target.

Phenotypic Assays

The historical standard method of antiviral susceptibility testing has been the PRA. It is the method to which new methods are usually compared. Many variations of the PRA have been reported; a standard exists as developed by CLSI (1). The principle of the PRA is the inhibition of viral plaque formation in the presence of viral agent. The concentration at which the antiviral agent inhibits plaque formation by 50% is referred to as the IC₅₀. Although the PRA is labor-intensive and uses reagents to a greater extent than other methods, it is appropriate for small-scale testing of isolates. As discussed earlier, titers of the viral isolate need to be adjusted to ensure an inoculum appropriate for the surface area of the testing unit, wells, or plates.

The DU assay, based on preferential uptake of a vital dye (neutral red) by living but not by nonviable cells, has been used mainly for testing HSV and will not be discussed in detail here.

DNA hybridization assays are semiquantitative measures of the amount of viral DNA produced in the absence and presence of antiviral drugs. The IC₅₀s are calculated from these end points. This approach has been used to measure the effect of different antiviral compounds on DNA synthesis. The procedure has been detailed and good correlation between the PRA and a dot blot hybridization assay has been demonstrated (6,7).

EIA susceptibility testing methods for various viruses including influenza A have been developed (8). EIAs permit quantitative measurement of viral activity by spectrophotometric analysis. The IC₅₀s are calculated as the concentrations of antiviral agents that reduce absorbance to 50% of the virus control. Susceptibility testing of influenza A virus by PRA is tedious; the EIA methodology is technically easier and is more suitable for testing multiple isolates. For influenza, the EIA approach uses antibodies to influenza A virus HAs (H1 or H3). Viral growth correlates with expression of viral HA.

Another approach for assessing antiviral activity is the use of YRAs. These assays reflect the ability of an antiviral agent to inhibit the production of infectious virus instead of the formation of a plaque. For testing antiinfluenza A compounds, cell monolayers are infected with virus in the presence and absence of antiviral compound and virus replication is determined by measuring HA titers. The test virus isolate is considered drug susceptible if the HA titer is reduced at least fourfold compared to untreated virus (9,10).

Genotypic Assays

The investigation of antiviral susceptibility or the converse resistance has been studied extensively for viral agents discussed in this chapter; more recently, these studies have addressed influenza viruses. Although not all mutations that cause resistance are known, many have been defined, permitting the application of genotypic and/or sequence analysis. Genotypic assays may be more costly than phenotypic approaches but their relatively rapid turnaround time compared to phenotypic assays confers a distinct advantage. Applicable assays will be discussed for each of the viruses.

Currently, aside from spontaneous mutations, most of the available antiviral compounds are able to induce mutations, at least at their active site, resulting in a modified viral replication capacity. This varies according to the mutations, some of which may be beneficial in reducing viral replication (viral fitness). However, reciprocal interaction between mutations can only be measured by phenotypic testing. Resistance testing is currently used in clinical practice for HIV infection. These mutations in proviral DNA are being investigated. For the other viral infections presented in this chapter, that is, HBV and HCV, cytomegalovirus (CMV), and influenza, sequencing procedures for clinical practice are still at the early stage and are typically restricted to treatment failure, but rarely for treatment choice, for lack of a significant number of available compounds. However, their use for epidemiologic purposes has been amply demonstrated. This has provided data on viral variability among different countries.

Considerable progress has been made in defining the indications for resistance testing and determining the cost-effectiveness of strategies that use testing in the management of HIV-infected individuals. Prospective, randomized trials have shown at least short-term virologic benefits for resistance testing (11). Moreover, emerging data indicate that viral drug resistance is a problem whenever treatment is used, and it may be increasing in importance. It has also become clear that knowledge concerning patterns of resistance and cross-resistance is critical to the development of successful sequential antiretroviral (ARV) regimens.

In developed countries, resistance testing has been adopted as the standard of care in case of failure to respond to ARV treatments for HIV infection (12,13). Drug resistance testing became an essential tool to assist clinicians in the selection of potent ARV drug regimens that will enhance the likelihood of favorable treatment responses. In the United States, more than 20 ARV drugs are approved for use; they belong to six classes of drugs based on their mechanism of action (Fig. 11.1). The ARV types are nucleoside/nucleotide reverse transcriptase inhibitors (NRTIs), nonnucleoside/nucleotide reverse transcriptase inhibitors (NNRTIs), protease inhibitors (PIs), chemokine receptor 5 (CCR5) antagonists, integrase strand transfer inhibitors (INSTIs), and fusion inhibitors.

To better understand the modes of action and antiviral resistance of HIV the basic structure of the virus should be appreciated. HIV is a member of the Lentivirus genus in the Retroviridae family. The retrovirus genome consists of two RNA molecules transmitted as a single stranded positive sense enveloped virus. Upon entry into the target cell, viral RNA is reverse transcribed into double-stranded DNA by virally encoded reverse transcriptase that is transported with the viral genome in the virus particle. Two different approaches are used to assess HIV drug resistance: genotyping and phenotyping. Understanding the characteristics, performances, and the interpretation of these assays is needed to use them optimally (14–16). Genotypic assays detect mutations expressed as nucleotide disarray resulting in amino acid changes that have been shown to correlate with in vitro and/or in vivo resistance to a particular drug or class of drugs. Phenotypic assays provide quantitative measure of drug susceptibility by determining the concentration of drug required to inhibit virus replication in cell culture.

ARV resistance due to viral gene mutations accounts for a large portion of treatment failures. The emergence of these genetic changes in HIV type 1 (HIV-1) is fostered by ongoing viral replication in the presence of subinhibitory concentrations of ARVs. Poor penetration of drugs into certain body compartments (sanctuary sites), inadequate adherence, and variable pharmacokinetic factors may contribute to subtherapeutic drug levels in vivo. This, in turn, may allow for selection of either preexisting (archived) or newly generated drug-resistant mutants. The critical problem in the clinical setting is that a mutant selected by a failing regimen may have some degree of cross-resistance to other drugs in the same class that have not yet been prescribed to that patient. The development of cross-resistance may lead to a reduced virologic or immunologic response to subsequent regimens. As scientists develop new agents active against resistant virus (see Fig. 11.1), clinical medicine is also implementing diagnostic strategies designed to detect ARV resistance and individualize subsequent regimens.

ARV resistance develops when viral replication continues in the presence of the selective pressure of drug exposure. As previously noted, the 20 approved ARV drugs belong to six classes based upon their mechanism of action (see Fig. 11.1).

For some drugs, such as the NRTI lamivudine and all available NNRTIs, a single mutation induces high-grade resistance in a predictable manner. For others such as zidovudine, abacavir, tenofovir, and most PIs, high-grade resistance requires the serial accumulation of multiple mutations and is thus slower to emerge.

Nucleoside and Nucleotide Reverse Transcriptase Inhibitors

Although most of the mutations associated with NRTI resistance are not at the active site of the enzyme, they do lead to conformational changes that affect the active site aspartate residues. Different mutations lead to two different mechanisms for resistance: decreased substrate binding and increased phosphorolysis (removal of the chain-terminating substrate that has already been incorporated into the growing proviral DNA chain). Both mechanisms lead to an overall net decrease in termination of the elongating chain of HIV DNA by the NRTI.

Nonnucleoside Reverse Transcriptase Inhibitors

Two patterns of multi-NNRTI resistance have been described. One is the K103N RT mutation. This single mutation confers resistance to all currently available NNRTIs, presumably by stabilizing the closed-pocket form of the enzyme, thus inhibiting the binding of the drug to its target. The fact that all available agents in this class bind to the same domain explains the broad pattern of cross-resistance and has prompted the development of new agents that interact with this domain more favorably. Indeed, another pattern of multi-NNRTI resistance is the accumulation of multiple mutations including L100I, V106A, Y181C, G190S/A, and M230L. Rarely, Y188L causes multi-NNRTI resistance.

Enhanced susceptibility to NNRTIs (i.e., hypersusceptibility) has been described in association with multiple mutations conferring broad cross-resistance to NRTIs and a lack of NNRTI resistance mutations. Longer duration of NRTI use, prior use of zidovudine, and abacavir or zidovudine resistance all have been associated with hypersusceptibility. This phenomenon appears to have biologic significance, with its presence enhancing the response to efavirenz-based regimens. A significantly greater short-term reduction in plasma HIV-1 RNA level, showing hypersusceptibility to efavirenz, was noted in patients who received that drug for salvage therapy.

Protease Inhibitors

The sequential use of certain PIs may be possible in some situations because several drugs in this class have distinctive major resistance mutations. This is particularly true for nelfinavir and has been suggested for atazanavir. All other PIs retain activity in vitro and in vivo against D30N isolates selected by nelfinavir. Less commonly, nelfinavir failure is associated with L90M, which is more likely to add to cross-resistance to other PIs. The I50V amprenavir resistance mutation alters the hydrophobic interaction with the target and had been thought to alter the binding of other drugs in this class only minimally. Clinical evidence to support particular PI sequencing, except that for nelfinavir, is lacking.

The presence of two key mutations (e.g., D30N, G48V, I50V, V82A/F/T/S, I84V, and L90M) generally confers broad cross-resistance to most currently available PIs. One strategy to avoid the accumulation of multiple mutations is to use low-dose ritonavir to increase the circulating levels (or “boost”) of other PIs (e.g., lopinavir, indinavir, amprenavir, and saquinavir), which may result in higher and more prolonged drug concentrations and greater suppression of viral variants that contain a limited number of mutations. Thus, resistance depends not only on intrinsic properties of the virus but also on the achievable plasma levels of the drug.

Fusion Inhibitors

The fusion inhibitor operates by interfering with the fusion of viral and cellular membrane and thus entry into the CD4 cell. It binds to the initial heptad repeat (HR1) in the gp41 subunit of the viral envelope glycoprotein and prevents conformational changes required for fusion of the viral and cellular membranes. Enfuvirtide is the currently available fusion inhibitor delivered by subcutaneous injection. This agent is typically reserved for salvage therapy.

Integrase Strand Transfer Inhibitors

ARVs of this category obstruct integrase by binding in the catalytic core domain of the enzyme and competing for binding with host DNA. This action prevents integrase from inserting the viral genome into host DNA. Raltegravir is the currently available compound. Dolutegravir is a once-daily alternative to raltegravir.

When initiating therapy for HIV-infected patients that are considered naive, that is, never before been treated for infection, the U.S. Food and Drug Administration (FDA) panel recommends that therapy consist of two NRTIs and at least one ARV from another ARV class including a NNRTI, INSTI, or PI. The reference Web site provides up-to-date preferred regimens and weighs the strength of recommendations (Panel on Antiretroviral Guidelines for Adults and Adolescents. Guidelines for the Use of Antiretroviral Agents in HIV-1-Infected Adults and Adolescents. http://www.aidinfo.nih.gov).

Entry Inhibitors

Entry of HIV-1 into target cells is a multistep process involving attachment (mediated by gp120 binding to CD4), chemokine coreceptor binding, and association of two trimeric helical coils (HR-1 and HR-2) located in the ectodomain of gp41 into a six-helix bundle that brings the virus and cell membranes into close approximation, allowing membrane fusion to occur. A number of drugs currently in development block HIV-1 infection by interfering with one of these steps. The recently approved fusion inhibitor enfuvirtide (known as T-20) blocks the association of HR-1 with HR-2 by binding to the trimeric HR-1 complex, thereby inhibiting fusion and blocking virus entry (17). Mutations in HR-1 that reduce enfuvirtide susceptibility are selected by in vitro passage of HIV-1 in the presence of the drug and have been identified in isolates obtained from patients receiving enfuvirtide in clinical trials.

Identification of the presence of drug resistance by means of genotypic or phenotypic resistance assays can help a health care provider select a combination of ARVs that is likely to suppress HIV-1 replication (i.e., “active drugs” to which that patient’s virus population is not cross-resistant). To maximize the therapeutic benefit and minimize toxicity, information collected from the viral genotype or phenotype must be used in conjunction with the patient’s ARV treatment history, response to past regimens, immunologic status, pharmacologic data, and the clinician’s own knowledge of ARV drugs. Knowing when and how to use resistance testing in a clinical practice will lead to better clinical management of HIV-1–infected patients.

Resistance assays use different technologies that provide complementary information about ARV resistance. As noted earlier, the two different types of drug resistance tests available are genotypic and phenotypic assays. Genotype assays provide information about viral mutations that may result in changes in viral susceptibility to particular drugs or classes of drugs. Phenotype assays directly quantitate the level of susceptibility of a patient’s virus sample to specific drugs in vitro. The values measured from the patient sample are compared with values measured from a standard wild-type reference strain. The degree of phenotypic resistance is the difference in susceptibility to a particular drug between the patient sample and the reference strain. In most cases, both genotype and phenotype testing methods require the use of polymerase chain reaction (PCR) technology to amplify the HIV-1 genes of interest (PR and RT) from patients’ plasma samples. However, there are numerous differences between these two resistance testing methods.

Phenotypic Testing

Several phenotypic assays have been in use for testing HIV-1. A serious limitation of some of these procedures is that not all clinical isolates grow in the cell culture lines used in these assays.

In the late 1980s, the first developed phenotypic assays were PBMC assays that required isolation of the virus by cocultivating patient’s PBMCs with mitogen-stimulated PBMCs obtained from HIV-seronegative donors, then titration of the virus stock. Subsequently, the inhibition of virus growth, in the presence of several concentrations of the drug, was evaluated by measurement of p24 Ag or RT activity in comparison with the replication in the absence of drug (18,19). These PBMC assay methods had several limitations regarding the difficulties in standardization, the interassay variability, and the burdensome workload for laboratory personnel. Overall, the PBMC compartment does not represent the actively replicating virus population present in the plasma compartment.

The development of RVA method, first described in 1994 (20), enabled the measurement of phenotypic resistance on a large scale. Since 1998, two companies have developed RVA assays: Antivirogram (Virco BVBA, Mechelen, Belgium) and PhenoSense (ViroLogic Inc, San Francisco, CA) (21). Antivirogram was the first RVA adapted to commercial development. The overall approach of the three commercial phenotypic assays is similar, but each assay is performed using different protocols (extraction, amplification) and reagents (viral vectors, cell lines, titration of the virus) (Table 11.1). Assays use PCR to amplify the entire protease, much of RT, and a part of the 3′ end of gag gene, including cleavage sites (p4/p2, p2/p7, p7/p1, p1/p6) from HIV-1 RNA extracted from patient plasma. Phenotypic PI resistance may be modulated by mutations at gag-pol cleavage sites, and four of the nine cleavage sites in the recombinant virus come from the patient virus and five from the laboratory virus construct. The amplified material is incorporated into vectors that derive from full-length molecular clones of HIV-1 but lack the protease and RT regions of the pol gene to create a recombinant HIV-1 isolate. After amplification, two strategies have been used to insert patient PR and RT sequences into vectors: The Antivirogram test (Virco BVBA, Mechelen, Belgium) uses homologous recombination following cotransfection of cell lines, whereas the PhenoSense assay (ViroLogic Inc, San Francisco, CA) uses site-specific endonuclease cleavage and direct ligation. Ligation products that are capable of propagating the vectors to high copy number are then introduced into bacterial cells. High virus stocks are generated by transfecting the recombinant viral vector DNA.

PhenoSense assay (ViroLogic Inc, San Francisco, CA) is a single-cycle assay using recombinant viruses that are limited to a single round of viral replication by a specific deletion in the HIV env gene of the vector. In this assay, evaluation of reverse transcriptase inhibitors (RTIs) involve serial dilutions of drugs added to the target cell line at the time of virus inoculation to block RT activity at the cell entry. PIs are evaluated by adding serial dilutions of drug to the transfected cells producing the virus stocks. In the Antivirogram (Virco BVBA, Mechelen, Belgium) assay, generation of high virus stocks is obtained by cultivating the recombinant viruses for several replication cycles (1 to 2 weeks). Then the virus stock is titrated before testing for drug susceptibility. Either PIs or RTIs can be evaluated using the same format by adding serial dilutions of drug to the cultures at the time of virus inoculation.

Commercially available RVA systems exploit the use of sensitive reporter genes that have been engineered into the target cells or the retroviral vector. Transcription of the reporter gene is placed under the regulatory control of the HIV-1 promoter/enhancer within the long terminal repeat (LTR). With infection with the recombinant virus, reporter gene expression is transactivated by the HIV-1 tat protein that is produced early in the HIV-1 replication cycle.

Drug susceptibility is assessed by comparing the IC₅₀ of the patient virus to the IC₅₀ of a drug-susceptible reference strain derived from the NL4–3 or HXB2 strain. Assay data are analyzed by plotting the percent inhibition of virus replication versus the log₁₀ drug concentration. The final results are expressed as fold change or resistance indices, which are calculated as the IC₅₀ of the patient virus divided by the IC₅₀ of the reference virus. The Antivirogram (Virco BVBA, Mechelen, Belgium) assay results report IC₅₀ values with graphic- and numerical-fold change values.

These phenotypic methods have been adapted to measure drug susceptibility to the entry inhibitors (22). The first results showed a very large variability in the IC₅₀ of isolates from drug-naive patients. Such variability of up to 3 logs difference in IC₅₀ is not yet explained.

Studies evaluating the comparative performances of the different phenotyping assays are limited. Excellent concordance was observed between the Antivirogram (Virco BVBA, Mechelen, Belgium) and PhenoSense (ViroLogic Inc, San Francisco, CA) assays (23) but the majority of the viruses tested were of the wild type.

Phenotypic Assays

Several phenotypic assays are available for determining the susceptibility of an HIV-1 isolate to nucleoside analog RT inhibitors. As noted earlier, a serious limitation of some of these procedures is that the clinical isolate recovered may not grow in the designated cell culture lines used in these assays. In order to meet this challenge, the AIDS Clinical Trials Group developed an assay performed in PBMCs that allows for most clinical isolates at HIV-1 (24). Viral end point is measured by quantitating p-24 antigens. To circumvent the typical drawbacks of phenotypic assays, a new iteration was developed, the RVAs. The two RVAs commercially available were previously noted: the Antivirogram (Virco BVBA, Mechelen, Belgium) assay (25) and the PhenoSense (ViroLogic Inc, San Francisco, CA) assay (21). In addition to the lengthy turnaround time of these phenotypic assays (~10 days) is the requirement for a minimum of 500 to 1,000 copies of HIV-1 RNA/mL plasma.

An unusual phenotypic assay exploits the requirement for HIV-1 to bind to a coreceptor, in addition to CD-4 for the virus to gain entry into the host cell; CXC chemokine receptor type 4 (CXCR4) and CCR5 are two major coreceptors for HIV-1 expressed on T lymphocytes and macrophages, respectively. A recently introduced entry inhibitor, Maraviroc, exploits the necessity for HIV-1 to bind to CCR5, preventing fusion of HIV-1 and T-lymphocyte membrane, thus blocking entry into cells. However, some HIV-1 strains exhibit “dual-tropism,” that is, they use both CXCR4 and CCR5 receptors; patient isolates must be characterized prior to initiation of Maraviroc therapy by means of a tropism assay to discern the required receptor(s). Commercially available tropism assay(s) are extant: the Phenoscript assay and the Trofile (Laboratory Corporation of America Holdings, Burlington, NC) assay. For integrase inhibitors such as raltegravir, an alternative specific phenotypic assay, PhenoSense Integrase (ViroLogic Inc, San Francisco, CA), is available. Mutations that confer resistance to integrase inhibitors have been elucidated and encompass two different pathways with a major mutation of either Q148/K/R or N155H. Using this information, laboratories can develop genotypic assays to detect this resistance capability.

Phenotypic drug susceptibility assay results are interpreted by comparing the replication of patient-derived viruses to replication of well-characterized laboratory strains at equivalent drug concentrations. The phenotypic assay cutoffs (change in IC₅₀) defining whether the patient viral strain is susceptible or resistant are evolving. Former assays used “technical” cutoffs that refer to the interassay variability of the controls (one- to twofold for PhenoSense [ViroLogic Inc, San Francisco, CA] and two- to fourfold for Antivirogram [Virco BVBA, Mechelen, Belgium]). The “biologic” cutoff was obtained by testing a large number of viruses from treatment-naive individuals and defining the natural distribution of drug susceptibility in HIV-1 strains. The most relevant interpretation of phenotypic assay results is based on “clinical” cutoffs, which are distinct for each ARV drug and different for each assay system. Clinically relevant cutoffs available for a limited number of drugs are based on analysis of the relationship between baseline phenotype and reduction in viral load (Table 11.2).

In the Antivirogram (Virco BVBA, Mechelen, Belgium) assay, biologic cutoff values are the basis for reporting assay results for all drugs tested except tenofovir and lopinavir/ritonavir, which are based on clinical response data. In the PhenoSense (ViroLogic Inc, San Francisco, CA) assay, results for lopinavir/ritonavir (26), abacavir (27), tenofovir (28), stavudine, and didanosine (29,30) are reported using clinically derived cutoffs. The PhenoSense (ViroLogic Inc, San Francisco, CA) assay can also measure increased susceptibility, often referred to as hypersusceptibility. This phenomenon has been mainly described for NNRTI drugs and is associated with resistance to nucleoside analogs (31). In a retrospective analysis of virologic outcome in patients who were NNRTI naive and NRTI experienced, increased susceptibility to efavirenz (as defined as less than 0.4-fold in susceptibility) was independently predictive of reduction in virologic failure (32). Even partial activity may be clinically useful when treatment options are limited (26).

There are ongoing efforts to develop clinical cutoffs for all approved ARV drugs, but some clinical cutoffs are difficult to establish as they are too close to the technical cutoff or the variability of the assay (33). When interpreting the results of phenotypic assay, the clinician should consider how each cutoff was derived (Table 11.2).

Genotypic Testing

The presence of resistance mutations by genotypic assays is identified by DNA sequencing or point mutation assays such as hybridization assays. Most diagnostic laboratories developed their own assays, which are referred to as home-brew assays. All genotypic assays require extraction of the virus genome, usually from the plasma specimen, then retrotranscription of the viral RNA in cDNA. Subsequently, the DNA is amplified in a single or nested PCR. Dideoxynucleotide sequencing is the standard approach to HIV genotyping (34).

Mutations in the nucleic acid sequence cause amino acid substitutions when messenger RNA (mRNA) is translated into protein. Some mutations are “silent mutations”; that is, the nucleic acid changes do not alter the amino acid sequence. Mutations are designated in a shorthand format using single-letter abbreviations for the amino acids encoded by a particular triplet of nucleotides (a codon). The normal, or wild-type, amino acid present at a particular location in a protein is given, followed by the location (amino acid position, or codon number), followed by the new amino acid that has replaced the wild-type amino acid. The designation L90M, for example, indicates that the amino acid methionine (M) has been substituted for the wild-type amino acid leucine (L) at position 90, which is one of the codons in the region that codes for the protease enzyme, and is designated PR. The region that codes for the reverse transcriptase enzyme is designated RT.

In 2001, a commercial HIV-1 RT and protease genotyping kit was approved by the FDA and European Medical Evaluation Agency (EMEA) for use in clinical settings. A second kit was FDA- and EMEA-approved in 2002 (Table 11.3). The advantages of using approved kits include standardization and consistency of results across laboratories, making them preferable to laboratory-developed tests (LDTs), previously known as home-brew tests in local laboratories with low experience in molecular biology. However, the commercial kits are more expensive and may not provide the flexibility of home-brew methods.

Overview of a Laboratory-Developed Test Sequencing Method: HIV-1 RNA Extraction, cDNA Synthesis, Amplification, and Sequencing

Genotypic Assays

The interactions between different mutations complicate the interpretation of genotypic assays. Mutations are designated by the wild-type amino acid present at a particular position of the RT or protease gene, followed by the amino acid position, then by the new amino acid that has replaced the wild-type amino acid. For example, M184V indicates that the methionine (M) wild-type amino acid at position 184 of the RT gene is replaced by the valine (V) amino acid. A large number of genotypic-resistance interpretation tools have been developed in recent years. These include mutation lists, rule-based algorithms, and interpretations based on databases correlating genotypes with corresponding phenotypic susceptibilities. The listing of mutations associated with resistance is available through the expert panel of International AIDS Society (IAS)-USA (http://www.iasusa.org) or Los Alamos National Laboratory (http://hiv-web.lanl.gov).

The interpretation of genotypic assays is based on interpretation systems called algorithms. They have the objective, in treated patients, to predict the response to each ARV according to the combination of mutations present on the RT and protease genes. More than 20 algorithms are available from different sources. Some algorithms are public and available on Web sites, such as the algorithms from Stanford University (http://hivdb.stanford.edu), the Rega Institute (http://www.kuleuven.ac.be), and the Agence Nationale de Recherches sur le Sida et les hépatites virales (http://www.hivfrenchresistance.org).

The early approach to interpreting genotypes was based on the in vitro correlation studies relating genotype with phenotype. However, recent data showed difficulty in determining reliable phenotypic cutoffs for some drugs such as stavudine, didanosine, and amprenavir (33,35). Correlation studies analyzing the virologic response in treatment-experienced patients according to the genotypic profile at baseline should provide the most relevant information for establishing algorithms. Such algorithms are still limited and available for some ARV drugs such as abacavir (36), stavudine (37,38), amprenavir (39), lopinavir (26,40,41) and tenofovir (28). The method issues for correlating baseline phenotype with virologic response also apply to analyses of baseline genotype and response. To build up such algorithms, a strict method is needed that is not standardized. Multivariate analyses must show the predictive value of the algorithm when there are confounding factors such as viral load at baseline, previous drug history, duration of past treatment, and new drugs in the regimen (42). Then the validation step must confirm that the algorithm is also predictive of the virological response in a different data set (37). These correlation studies are based on retrospective analyses of patients enrolled in therapeutic trials that are sometimes several years old. The accuracy of these algorithms depends on the prevalence of specific resistant mutations at baseline; some mutations, relevant to the resistance but underrepresented in the genotype profile, will be ignored by the algorithm.

No single study can be expected to provide a full picture of the relationship between genotype and response. The genotypic profile of patients is changing as new drugs become available that may select for previously unknown mutations. There is a need for wide-ranging databases containing appropriately quality-controlled data from genotypic resistance assays and international efforts to be developed to pool databases and to establish standardized analyses for constructing algorithms that need to be frequently refined.

Different reporting formats and interpretation systems are provided by the wide diversity of clinical virology laboratories providing HIV genotyping. Laboratories using home-brew assays have varied approaches to reporting the results. Some laboratories provide only a list of mutations, and understanding by clinicians is poor. In addition to listing each resistance mutation detected in RT and PR genes, the interpretation must be provided with the precise indication of the algorithm used.

Reports of genotypic resistance usually score virus isolates as “resistant,” “possible resistant,” or “no evidence of resistance.” Some algorithms classify patient isolates in four or five categories. Possible resistance corresponds to different situations according to the algorithm and the drugs: The detected mutations may have been associated with diminished virologic response in some but not all patients. This classification may also refer to a limited knowledge on resistance to this particular drug.

The use of expert advice in interpreting a genotype result has been shown to lead to a better choice in the alternative regimen compared with a regimen chosen in the absence of expert advice (43). Unfortunately, such expert advice is not always available. The role of the expert is to know the bases of the algorithm and to modulate the interpretation according to the clinical and immunologic parameters of the patient.

The discordances in interpretations seem to be drug-related, low for NNRTI, and more important for some ARVs such as stavudine, didanosine, abacavir, and amprenavir (44,45). Several studies retrospectively analyzed the relationship between the different interpretations of resistance genotypes by several algorithms and the outcome of salvage treatment in cohorts of drug-experienced patients (45,46). These studies show generally significant discordances between available algorithms that are associated with different predictions of subsequent virologic outcomes.

Virtual Phenotype

Another approach to interpreting genotype is employed by Virco (47). This company maintained a large proprietary database of specimens for which both genotype and phenotype are known for each drug. From this database, viruses with the same genotype as that of the patient’s virus are identified and the average IC₅₀ of these matching viruses is calculated, estimating the likely phenotype of the patient’s virus. Several reports support the validity of the virtual phenotype for interpreting genotypes (48). However, the accuracy of the predicted phenotype depends on the number of database matches with the patient’s isolate. For the new or investigational agents, the system is delayed by the need to accumulate genotype–phenotype data; with the availability of new drugs and the selection of new mutations, the number of matches may decrease.

Appropriate interpretation of the results of drug resistance testing is a challenging problem for both phenotype and genotype assays (12). These interpretation tools show variability, which may modulate the choice of drugs to be used and therefore affect the therapeutic outcome. More attention is being focused on the importance of a standardized approach to interpret resistance results. Interpretation of resistance testing and choice of new therapy must be performed in light of all clinical information, including past therapies, previous viral load and immunologic responses, adherence, tolerance, and toxicities. Previous resistance test results must be considered when available.

Several studies comparing phenotype and genotype results on a large number of clinical samples showed that discordances between results are not uncommon (49,50). The clinical usefulness of genotypic testing has been demonstrated in most prospective, randomized studies; in contrast, phenotypic testing has been shown to be clinically useful in a few prospective studies. Genotypic testing is also used more commonly than phenotypic testing because of its lower cost, wider availability, and shorter turnaround time. Unlike genotyping assays, phenotyping assays involving sophisticated laboratory procedures are not available as test kits for widespread distribution.

Genotypic testing may detect a single drug resistance mutation within a virus population that will affect the virologic response but will not reduce phenotypic drug susceptibility. For example, the Y181C mutation confers in vitro susceptibility to efavirenz (51) but is associated with absence of clinical response to this drug (52). Genotypic assays detect mutations present as mixtures even if the mutation is present at a level that is too low to affect drug susceptibility in a phenotypic assay and identify reversal mutations that do not cause phenotypic drug resistance but indicate the presence of previous drug pressure. Decreased phenotypic susceptibility to some drugs may be suppressed (resensitization) by other mutations in the sequence. In the presence of thymidine mutations that decrease susceptibility to zidovudine, the presence of M184V or L74V or K65R may restore the in vitro susceptibility to this drug. Such in vitro–increased susceptibility has not been proved to be clinically relevant.

Phenotypic testing in research settings is essential for establishing genotype-phenotype correlations, which provide the first bases for interpreting genotype tests and for designing new antiviral drugs that are effective against existing drug resistance strains.

Relationship of Genotype and Phenotype Results to Drug Levels

Virus drug susceptibility is likely to be a continuous phenomenon because of partial remaining activity of the drugs against mutant viruses and variability of drug exposure.

Data correlating drug concentrations with virologic response have been generated for most PIs and NNRTIs. The concept of inhibitory quotients (IQs) characterizes the relationship between drug exposure and drug susceptibility of the virus and is defined as the C_min of drug divided by measure of resistance either by phenotype (IQ: C_min/IC₅₀), “virtual” phenotype (vIQ: C_min/value of IC₅₀ according to the virtual phenotype), or genotype (GIQ: C_min/number of mutations). This concept has begun to be applied to prediction of response to PIs (53). The concept is particularly suitable for PIs that exhibit very large interindividual drug concentration variability. Although IQs have yet to be prospectively evaluated as a tool for managing HIV infection, they have been shown to be better predictors of virologic response in treatment-experienced patients receiving lopinavir boosted by ritonavir than plasma drug concentrations and/or resistance testing alone (53). The virtual IQ was the best predictor of viral load reduction in response to ritonavir boosting indinavir-based therapy in patients with ongoing viremia (54). The accurate adjustment of the in vitro–calculated IC₅₀ to the in vivo protein binding remains to be determined. In PI-experienced patients receiving boosted amprenavir, the genotype IQ, using the C_min measured at week 8 and the number of PI resistance mutations evaluated at baseline, was the best predictor of the virologic response at week 12 (55). These approaches need to be validated in prospective clinical trials.

Practical Considerations

All genotypic and recombinant phenotypic assays use initial amplification through reverse transcriptase polymerase chain reaction (RT-PCR) as the first step in the process. Because these assays require amplification of a larger segment of the HIV-1 genome than assays designed only to detect the presence of HIV-1 RNA, they are generally less sensitive than viral load assays and require samples with at least 1,000 copies per milliliter of HIV-1 RNA to obtain amplification from plasma, although amplification is possible in some samples at lower viral load.

Plasma is the main source of virus used for testing HIV-1 drug resistance in the clinical setting. Because the half-life of HIV-1 in plasma is approximately 6 hours, only actively replicating virus can be isolated from this source; thus, the sequence of plasma virus represents the quasispecies most recently selected by ARV drugs (56). Specialized testing for research purposes can use a variety of other tissue compartments such as cerebrospinal fluid, genital secretions, PBMCs, or lymph nodes.

Blood samples may be drawn in either EDTA or acid citrate dextrose vacutainers. Heparin must be avoided as it inhibits PCR reactions. Plasma separation should be performed within a maximum of 6 hours after blood collection. Sample volume consists of 1 to 3 mL of plasma that can be stored at −80°C. Samples taken for viral load should be stored frozen within the laboratory in order that retrospective resistance testing can be undertaken.

These tests are technically demanding, and external quality control is essential. This is addressed by national and international pathology laboratory accreditation programs. Laboratories undertaking resistance testing should provide clinical support to HIV clinics and demonstrate participation in external quality control programs and accreditation by national and international agencies. In addition to quality assurance of the assay, quality assurance of the laboratory performing the assay is also required. Currently, all the laboratories in the United States that perform genotyping or phenotyping assays must have certification according to the Clinical Laboratory Improvement Act (CLIA) 1988 indicating some level of review of the laboratory’s performance standard (CLIA Related Federal Register, FDA, 1995. Available at http://www.fda.gov/cdrh/clia/fr/hsq230n.html). In Europe, some hospital laboratories participate in proficiency testing programs for genotyping.

Drug costs are driving the overall total cost of HIV-1 care in developed countries. A National Institutes of Allergy and Infectious Diseases and Centers for Disease Control and Prevention (CDC)–funded study analyzed cost-effectiveness in genotypic resistance testing using an HIV-1 stimulation model of 1 million patients (57). The authors reported that the cost-effectiveness of genotypic resistance is similar or better than that of recommended interventions for HIV-1-infected patients, such as Mycobacterium avium complex prophylaxis.

There is large geographical variability in reimbursement of HIV drug resistance assays. Lack of or low levels of reimbursement may still limit access in some areas or countries. However, proper implementation of resistance assays may reduce the overall cost of patient management by prompting more appropriate choices in therapy and avoiding drugs that are likely to induce toxicities.

Quasispecies

Resistance testing has technical limitations. Both genotypic and phenotypic testing depends on PCR amplification of virus from plasma and therefore do not address the properties of different virus components (i.e., whole virus vs. viral genome alone). The likelihood of generating sufficient genome product to undertake further analysis depends on the starting concentration of virus. The nature of direct PCR sequencing techniques limits the detection of minority strains of virus within the plasma virus population to 20% (17,58–61). Smaller proportions of mutant virus may contribute to subsequent therapy failure and will not be detected. This component of the variable-limited portion of the virus population contributes to the concept of “quasispecies.” This limitation is particularly troublesome in patients with complicated treatment histories or in those who have discontinued one or more ARV drugs. To maximize the likelihood of identifying drug resistance mutations present within the virus population of a patient, it is important to analyze plasma samples for resistance testing before changing or discontinuing ARV therapy and to consider the patient’s treatment history when interpreting the results of resistance testing.

Direct PCR is done in clinical settings because it is quicker and more affordable than testing multiple clones individually. Clonal sequencing of individual strains is performed in research settings to answer questions about the pathways and evolution of HIV-1 drug resistance. Multiple quasispecies with distinct resistance genotypes coexist at any given time and some initially minority populations, with or without additional changes, can subsequently emerge as majority populations (62,63).

Cellular Reservoirs

The evolution of HIV-1 drug resistance mutations in proviral DNA in PMBCs lags behind that in plasma HIV-1 RNA. In patients with multiple virologic failures, proviral DNA may contain multiple archived mutations that are not present in plasma (64–67). Discrepancies in protease and RT resistance profiles have been described between plasma and other compartments such as cerebrospinal fluid or semen. The variable penetration of ARV drugs into sanctuary sites may contribute to the differential evolution of HIV and the emergence of drug resistance (68–70). However, the utility of sequencing virus from PBMC or from other sanctuary sites has not been evaluated in either prospective or retrospective clinical trials.

Non-B Subtypes

HIV-1 group M has evolved into multiple subtypes that differ from one another by 10% to 30% along their genomes. In North America and Europe, most HIV-1 isolates belong to subtype B. However, subtype B accounts for only a small proportion of HIV-1 isolates worldwide and non-B isolates are being identified with increasing frequency in Europe. Technically, primers used for RT, PCR, and sequencing may have a lower rate of annealing for non-B compared with subtype B templates. The performances of the different phenotypic and genotypic assays are currently being investigated. The ViroSeq HIV-1 genotyping system (Celera, Alameda, CA) was used to determine protease and RT sequences from a panel of 126 non-B subtypes isolates. Four specimens that could not be amplified included three subtype D isolates and one CRF02-AG isolate (71). The TruGene assay (Visible Genetics Inc, Toronto, Canada) using prototype 1.5 RT-PCR primers and the ViroSeq (Celera, Alameda, CA) assay were both successful for sequencing 34 non-B isolates, although five isolates (two belonging to subtype C, one to subtype B, one to subtype E, and one to subtype H) lacked double-strand sequence coverage in the ViroSeq assay (Celera, Alameda, CA) (72,73).

When mutation sequences were studied in subtype B and non–subtype B population, it was determined that all known subtype B resistance mutations were extant in the non-B subtypes and 80% were correlated with ARV treatment of patients with the non-B subtypes (74). Based on this conclusion, it appears reasonable to continue to focus on known subtype B resistance mutations for global monitoring of resistance.

It should be recognized that in general genotypic assays regardless of the methodologic approach can only detect known resistance-associated mutations. A concern with each genotypic methodology is that mutant variants may be present in the patient’s infection at low frequencies and may not be detectable so that mixtures of HIV-1 strains with minor sequence variations may not be distinguishable.

As resistance to ARVs is anticipated to be ongoing, there will be pressure to develop new drugs and approaches to therapy for HIV infection. Clearly, the testing methodology will be genotypic, with whole genome sequencing of the virus and interpretation and resistance testing being a key component.

Hepatitis B virus (HBV) was the first of the hepatitis viruses to be discovered. Whereas infection during adulthood is frequently cleared, vertical transmission from mother to child leads to persistent infection. More than 350 million people worldwide are currently persistently infected with HBV and are at risk of developing liver cirrhosis and hepatocellular carcinoma (75).

The HBV is a noncytopathic, parenteral DNA virus. The outcome of HBV infection depends on the kinetics of the virus–host interaction and particularly on the strength of the innate and adaptive, humoral, and cellular immune response. Whereas patients with acute hepatitis B have a vigorous and polyclonal immune response to HBV antigens, individuals with chronic hepatitis B have a weak and restricted immune response to HBV (76). Studies demonstrated that the latent, immune-mediated clearance mechanisms become activated spontaneously in chronically infected individuals undergoing hepatitis B e antigen (HBeAg) clearance (77). In addition, it has been shown that immune clearance of HBV can occur via cytolytic as well as noncytolytic mechanisms (78).

The asymmetric replication of the HBV genome, via RT of an RNA intermediate, makes it prone to mutations. Although mutations can occur randomly along the HBV genome, the overlapping open reading frames (ORFs) limit the number and location of viable mutations. Viable variants are selected on the basis of replication competence of the virus, selection pressure from the host’s immune system, and in some instances, exogenous factors such as antiviral therapy. The divergence of HBV sequences results in HBV quasispecies, variants, and genotypes with epidemiologic and clinical significance.

Structure of Hepatitis B

Virions

The HBV is an enveloped virus containing 3.2-kb, partially double-stranded DNA contained in a relaxed circular genome; it is the prototype member of the family Hepadnaviridae (79), which includes genetically similar viruses that infect primates and monkeys, woodchucks and ground squirrels, and herons and ducks. Electron microscopic examination of the serum of a highly viremic carrier reveals three types of virus-associated particles. The HBV virion, which is 42 nm in diameter, comprises an outer envelope formed by the hepatitis B surface antigen (HBsAg). This envelope surrounds an inner nucleocapsid made up of the hepatitis B core antigen (HBcAg) that packages the viral genome and associated polymerase. Abundant spherical particles 17 to 25 nm in diameter in numbers up to 10¹³ per milliliter and less numerous tubular structures or filaments approximately 20 to 22 nm in diameter and of variable length in numbers up to 10¹¹ per milliliter do not contain HBV DNA and thus are not infectious.

Viral Genotypes

There are currently seven recognized genotypes of HBV, designated A to G, that vary by 8% at the nucleotide level over the entire genome (80). The HBV genotype designation is based on the entire genomic sequence and thus is more reliable than the serologic subtype nomenclature that was used previously, which was based on the immunoreactivity of particular antibodies to a limited number of amino acids in the envelope protein. The relationship between the four major subtypes (adw, adr, ayw, and ayr) and genotypes has been determined (Table 11.4) (81).

Important pathogenic and therapeutic differences do exist among HBV genotypes (82). For example, genotype C is associated with more severe disease than genotype B, and genotype D is associated with more severe disease than genotype A (83). Genotypes C and D are associated with a lower response rate to interferon therapy than genotypes B and A (81,83). Recombination between two HBV genotypes has been reported for genotypes B and C (84) and genotypes A and D (85), generating more diversity.

Genome and Common Mutants of Hepatitis B Virus

The genome of HBV is a partially double-stranded, relaxed, circular DNA molecule about 3,200 nucleotides (nt) in length (Fig. 11.2) (79). The two linear DNA strands are held in a circular configuration by a 226-base pair (bp) cohesive overlap between the 5′ ends of the two DNA strands that contain 12-nt direct repeats called DR1 and DR2 (79). All known complete HBV genomes are gapped and circular, comprising between 3,181 and 3,221 bases, depending on the genotype (see Table 11.4). Within the virion, the minus strand has a fixed length with defined 5′ and 3′ ends and a terminal redundancy of 8 to 9 nt (86). The minus strand is not a closed circle and has a nick near the 5′ end of the plus strand. The viral polymerase is covalently bound to the 5′ end of the minus strand. The 5′ end of the plus strand contains an 18-base-long oligoribonucleotide, which is capped in the same manner as mRNA (86). The 3′ end of the plus strand is not at a fixed position, so most viral genomes contain a single-stranded gap region of variable length ranging from 20% to 80% of genomic length that can be filled in by the endogenous viral DNA polymerase.

The minus strand contains four ORFs and carries all the protein-coding capacity of the virus (79). Importantly, these overlap in a frame-shifted manner with one another so that the minus strand is read one and a half times (79). The longest ORF encodes the viral polymerase (POL). The ORF for the envelope (PreS/S) genes is completely located within the POL ORF, and the ORF for the core (PreC/C) and X genes partially overlap with POL ORF. Thus, the HBV encodes more than one protein from one ORF by using multiple internal AUG codons within an ORF, creating additional start sites for protein biosynthesis. Nested sets of proteins with different N-termini are thus synthesized (80).

The HBV mutation frequency has been estimated to be approximately 1.4 to 3.2 × 10⁻⁵ nt substitutions per site per year, approximately 10-fold higher than for other DNA viruses (87). The HBV POL is an RT and lacks proofreading function. The mutation rate of HBV is also influenced by the clinical phase of the patients, such as immune tolerance versus immune elimination, and clinical settings such as immunosuppression and transplantation (88). The predominant HBV exists in an infected individual as the major population of the HBV quasispecies pool. The stability of the predominant HBV within this pool is maintained by particular selection pressures from the host’s immune system and the constraints imposed by the overlap in reading frames, viability, and replication competence of the virus. Furthermore, the magnitude and rate of virus replication are important, with the total viral load in serum frequently approaching 10¹¹ virions per milliliter. Most estimates place the mean half-life of the serum HBV pool at about 1 to 2 days, so that the daily rate of the de novo HBV production may be as great as 10¹¹ virions.

The high viral loads and turnover rates, coupled with poor replication fidelity, all influence mutation generation and the extent of the HBV quasispecies pool. Furthermore, the availability of “replication space” requires that the eventual takeover by a mutant virus depends on the loss of the original wild-type virus, which is itself governed by factors such as replication fitness and the turnover and proliferation of hepatocytes (89,90). Replication space can be understood in terms of the potential of the liver to accommodate new HBV covalently closed circular DNA molecules (cccDNA; Fig. 11.3). Synthesis of new cccDNA molecules can occur only if uninfected cells are generated by growth within the liver, hepatocyte turnover, or loss of cccDNA from existing infected hepatocytes (91,92). Thus, the expansion of a (drug-) resistant mutant in the infected liver can be possible only with the creation of new replication space.

Mutations in the Precore/Core Promoter, Precore, and Core Genes

Two major groups of mutations have been identified that result in reduced or blocked HBeAg expression. The first group includes a translational stop codon mutation at nt 1896 of the precore gene (93). The second group of mutations affect the precore/core (pre-C/C) promoter, also called basic core promoter, at nt 1762 and nt 1764, resulting in a transcriptional reduction of the pre-C/C mRNA (Table 11.5).

The single-base substitution (G-to-A) at nt 1896 gives rise to a translational stop codon (TGG to TAG) in the second last codon (codon 28) of the precore gene located in the epsilon (ε) structure. The ε structure is a highly conserved stem loop structure, with the nt G1896 forming a base pair with nt 1858 at the base of the stem loop. In HBV genotypes B, D, E, G, and some strains of genotype C, the nt 1858 is a thymidine (T). Thus, the stop codon mutation created by G1896A (T-A) stabilizes the ε structure. In contrast, the precore stop codon mutation is rarely detected in HBV genotype A, F, and some strains of HBV genotype C, as the nt at position 1858 is a cytidine (C), maintaining the preferred Watson-Crick (G-C) base pairing.

Three other mutations (at nt positions 1817, 1874, and 1897) that cause truncations in HBeAg have been reported (88). In addition, changes that affect the initiation codon at nt 1814, 1815, and 1816 have been described. The mutation at G1899A is frequently detected in association with the precore stop mutation of G1896A. Early studies implicated the HBV precore stop codon mutant, leading to HBeAg negativity, as a possible virulence factor for severe liver disease and fulminant hepatitis B (94,95). However, this strain has also been found in asymptomatic carriers (96).

The pre-C/C promoter mutations, such as A1762T plus G1764A, may be found in isolation or conjunction with precore mutations, depending on the genotype (see Table 11.5). The double mutation of A1762T plus G1764A results in a decrease in HBeAg production of up to 70% (97). This mutant strain display reduced binding of liver-specific transcription factors, resulting in less pre-C/C mRNA transcripts and thus less precore protein. However, this mutation does not affect the transcription of pregenomic RNA (pgRNA) or the translation of the core or polymerase protein. Thus, by removing the inhibitory effect of the precore protein on HBV replication, the pre-C/C promoter mutations appear to enhance viral replication by suppressing pre-C/C mRNA relative to pgRNA (97). As with the precore mutations, the pre-C/C promoter mutations have not been conclusively identified as a potential virulence marker.

The core gene possesses both B cell and cytotoxic T-lymphocyte epitopes, and for the virus to persist in the infected host, during the elimination phase of hepatitis B, escape mutations within those epitopes are readily selected (88,97). Akarca and Lok (98) have demonstrated that the frequency of core gene mutations is associated with the presence of precore stop codon mutations, HBeAg negativity, and active liver disease.

Mutations in the X Gene

Mutations in the X region can involve the regulatory elements that control replication such as the pre-C/C promoter and the enhancer 1. Because the pre-C/C promoter encompasses nt 1742 to 1802 and overlaps with the X gene in the concomitant reading frame, the A1762T plus G1764A pre-C/C promoter mutations also change in the X gene at xK130M and xV131I. A novel class of HBx mutants has been found in patients with hepatocarcinoma exhibiting increased clonal outgrowth and decreased apoptosis, implying a possible role in hepatocarcinogenesis (99).

Mutations in the Envelope Gene

Most hepatitis B vaccines contain the major HBsAg, and an immune response to the major hydrophilic region induces protective immunity. Mutations in this epitope have appeared under pressure generated by vaccine-induced antibodies (100). In addition, mutations have been detected after treatment of liver-transplant patients with hepatitis B immunoglobulin (101). Most isolates have a mutation from glycine to arginine at residue 145 of HBsAg (sG145R) or aspartate to alanine at residue 144 (sD144A) (see Table 11.5). The former mutation has been shown to evade the known protective anti-HBs response.

Viral Life Cycle

An understanding of the HBV life cycle is crucial for the identification of potential antiviral targets (102,103). HBV replication begins when the virion attaches to an as yet unidentified receptor on the hepatocyte surface (see Fig. 11.3). Following viral entry, the virus uncoats and is transported to the nucleus where the relaxed circular genome is converted by the host cellular machinery to the cccDNA; the cccDNA is, in turn, organized into viral minichromosomes. This key replicative intermediate is the transcriptional template for production of the various HBV RNAs, including the pgRNA, that are necessary for viral replication and represents one of the major obstacles in the development of effective treatments for the control of HBV infection. Transcription of the 3.5-kb pgRNA serves three important roles. First, its translation leads to production of the core and POL proteins. Second, it participates in the nucleocapsid packaging reaction, the specificity of which is provided by a unique stem-bulge-stem structure known as epsilon (ε) on the 5′ and 3′ ends of the pgRNA (104). Following translation of the pgRNA, the POL protein binds to the 5′-end ε. Host proteins such as heat shock protein 90 (Hsp90) stabilize this POL–ε interaction (104). The cis-translated core proteins dimerize around the pgRNA-pol complex and self-assemble to form viral nucleocapsids (105).

Once packaged into a nucleocapsid, the pgRNA serves its third and most important role as the template for reverse transcription and DNA synthesis. The POL protein is bound to the 5′ ε structure and acts as its known primer for initiation and synthesis of the first three nucleotides of the negative-strand DNA (102,103). This nascent DNA is then translocated to the 3′ end of the pgRNA where it binds to the complementary sequence within a 12-nt region known as direct repeat 1 (DR1). From here, negative-strand DNA synthesis proceeds and the RNase H activity of the POL protein degrades all but the last few nucleotides of the template pgRNA. This RNA oligomer is translocated to the 3′ copy of the 12-nt repeat known as DR2, from which point positive-strand DNA synthesis begins. Elongation of the positive-strand DNA proceeds to the 5′ end of the negative strand where a third strand transfer occurs. This is facilitated by a short redundancy (r) in the negative-strand DNA template, which anneals to the r region on the 5′ end positive-strand DNA, thereby circularizing the genome (102,103). Premature termination of the positive-strand DNA synthesis by the POL protein results in the characteristic partially double-stranded genome. The HBV nucleocapsid containing the partial double-strand DNA is either recycled back to the nucleus to increase the supply of cccDNA, or undergoes further processing in the endoplasmic reticulum and Golgi for virion assembly. Mature virions are subsequently exported from the cell via the constitutive secretory pathway.

Antiviral Drugs

Success in HIV drug development in the 1990s revolutionized treatment of hepatitis B. Several antiviral agents that were developed for the treatment of HIV infection proved to be effective in inhibiting HBV replication. Several therapeutic interventions are now available encompassing five nucleoside analogs: 2′,3′-dideoxy-3′-thiacytidine (3TC), adefovir dipivoxil, telbivudine, tenofovir, and entecavir. All treatments have limited long-term efficacy. Although there have been few direct comparison trials, the short-term efficacy for both HBeAg-positive and HBeAg-negative hepatitis B appears comparable; factors predictive of response are similar. Thus, the advantages and disadvantages of each treatment, the durability of response, and the patient’s preference must be carefully weighed before a decision is made (106).

The review of the HBV life cycle reveals that apart from reverse transcription, most viral processes depend on host cell machinery. The most important of these is the generation and persistence of cccDNA. Conventional antiviral inhibitors of viral DNA synthesis such as nucleoside/nucleotide analogs (Table 11.6) can prevent or reduce the development of new molecules of cccDNA. However, successful elimination of the existing pool of hepadnaviral cccDNA has only been achieved by either a noncytolytic T helper type 1 (Th1) immune response (107,108) or immune-mediated cell killing followed by hepatocyte division (109,110). In this context, it is important to note that treatment of hepatitis B with nucleoside analogs can result in the partial restoration of specific immunoresponsiveness, which appears necessary for durable host-mediated control of infection (111). Thus, the concept of successful therapy for hepatitis B is converging on the use of both antiviral and immunomodulating approaches. In this chapter, the focus is on antiviral drugs.

Evaluation of Drug Efficacy

The major role of serum HBV DNA assays in patients with hepatitis B is to assess HBV replication and candidacy for antiviral therapy. Quantitative HBV DNA testing in serum is also important in assessing the response to antiviral treatment. Historically, the molecular assays for HBV DNA detection and quantification were not well standardized. However, over the past several years with the onset of real-time PCR, there have been significant advances. Several commercial “cleared” (approved) assays using real-time PCR or transcription-mediated amplification (TMA) from five different manufacturers have become available. The lower limit of detection (LLOD) ranges from 6 to 30 IU/mL of HBV with the upper end of the dynamic range at 8.0 log₁₀ IU/mL. These assays have been reviewed by Valsamakis (112) and more recently by Chevaliez and colleagues (113) detailing LLODs and dynamic ranges.

A National Institutes of Health workshop on the management of hepatitis B proposed that the definitions and criteria of the response to antiviral therapy of hepatitis B be standardized. An arbitrary value of more than 10⁵ viral copies per milliliter has been chosen as a diagnostic criterion for hepatitis B (106). However, this definition is not perfect. The proposal categorized responses as biochemical, virologic or histologic, and as on-therapy or sustained off-therapy (106) (Table 11.7).

As for HIV-1, both phenotypic and genotypic assays have been used to characterize HBV drug resistance mutations. Various phenotypic methods have been studied; several are reviewed by Shaw et al. (114). In methods involving the development of point mutations associated with drug resistance, the phenotype is deduced by comparing replications of cell lines with or without mutations in the presence/absence of drugs. As with HIV-1, virtual phenotyping has been established that correlates patient clinical results with viral mutational data to assign a viral phenotype (114). Known mutations associated with resistance to HBV antiviral compounds are discussed by Chao and Hu (115).

Lamivudine and Virogram

Hepadnaviruses replicate through an RNA template that requires RT activity. HBV DNA polymerase was shown to share homologies with the RT from retroviruses. Inhibitors for RT of oncogenic RNA viruses may suppress HBV DNA replication. Lamivudine (Epivir) is the minus enantiomer of 3TC. It was developed as an RTI for use in HIV infection. Lamivudine has been shown to be a potent inhibitor of HBV replication in 2.2.15 cells (116). The 2.2.15 cells were clonal cells derived from Hep G2 cells that were transfected with a plasmid containing HBV DNA. These cells secreted hepatitis B virions. The 2.2.15 cells were maintained in minimal essential medium supplemented with 10% (vol/vol) fetal bovine serum (FBS). Cells were incubated at 37°C in a moist atmosphere containing 5% CO₂/95% air. The 2.2.15 cells were inoculated at a density of 3 × 10⁵ cells per 5 mL in a 25-cm² flask.

The drugs studied were added to the medium 3 days after the inoculation. Cells were grown in the presence of drugs for 12 days with changes of medium every 3 days. After incubation, the medium was centrifuged (10 minutes at 2,000 × g) and polyethylene glycol (M_r, 8,000) was added to the supernatant to a final concentration of 10% (wt/vol). The virus was pelleted (10 minutes at 10,000 × g). The pellet was resuspended at 1% the original volume in TNE (10 mM Tris HCl, pH 7.5/100 mM NaCl/1 mM EDTA). The suspension was adjusted to 1% sodium dodecyl sulfate (SDS) and proteinase K at 0.5 mg/mL and incubated for 2 hours at 55°C. The digest was extracted with phenol/chloroform, 1:1 (vol/vol), and the DNA was precipitated with ethanol. The DNA pellet was dissolved in TE₈₀ (10 mM Tris-HCl, pH 8.0/1 mM EDTA) and then electrophoresed in a 0.8% agarose gel followed by blotting onto Hybond-N membrane. The blot was hybridized with a ³²P-labeled HBV DNA (Bam HI insert from plasmid pam6; American Type Culture Collection [ATCC]) probe, washed twice with standard saline citrate (SSC)/0.2% SDS at room temperature for 1 hour and 0.1 × SSC/0.2% SDS at 55°C for 30 minutes, and then autoradiographed. The intensity of the autoradiographic bands was quantified by a scanning densitometer. HBID₅₀ was defined as the drug concentration that inhibited HBV viral DNA yield in the medium by 50%. The values were obtained by plotting percentage inhibition compared with control versus the drug concentration.

No cell growth retardation or effects on mtDNA was observed after the administration of lamivudine at concentrations at least 100 times higher than concentrations that completely block HBV replication. Lamivudine did not affect the integrated HBV DNAs. Since the RNA replicative intermediates are transcribed from the integrated DNA, it is not surprising that HBV-specific transcripts were not affected by drug treatment. Thus, interruption of drug treatment resulted in a return of HBV virus to intra- and extracellular populations (116). Lamivudine is phosphorylated in vivo to the active triphosphate (3TC-TP) that competes with deoxycytidine triphosphate (dCTP) for DNA synthesis. It inhibits DNA synthesis by terminating the nascent proviral DNA chain.

Emergence of the YMDD mutation was associated with duration of therapy. Sixty-seven percent of patients who received lamivudine for 4 years showed evidence of the YMDD mutation, 40% of patients showing evidence after 104 weeks, 31% of patients showing evidence between weeks 52 and 104, and 14% showing evidence after 52 weeks (117,118). Drug resistance can be detected as early as 49 days after taking lamivudine (119), but clinical evidence (phenotypic expression) of drug resistance, which is indicated by a rise in serum alanine transferase (ALT), does not occur before 6 months. Mutations in the catalytic polymerase/RT domain of the gene for HBV polymerase have been associated with lamivudine resistance in patients receiving treatment for HBV infection (120). This is one of the four functional domains of HBV polymerase, which also possesses a priming region, a spacer domain, and a region with ribonuclease H activity (91). The identification of several regions of conserved sequences within the polymerase/RT domain has led to its further subdivision into subdomains A to F. A sequence of four amino acids within the C subdomain consisting of tyrosine (Y) followed by methionine (M) and two aspartic acid (D) residues is highly conserved among viral polymerase/RTs. Termed the YMDD motif, it is essential for polymerase activity because of its involvement in binding nucleotide substrates in the catalytic site (120). Lamivudine resistance has been associated with substitution of isoleucine (I) or valine (V) for methionine in the YMDD motif at position 552 (rtM204I/V) of HBV polymerase/RT (see Table 11.5) (121). In vitro assays have confirmed that the YMDD motif mutations conferred reduced susceptibility of HBV to lamivudine (120,122). However, in vitro experiments also suggested that YMDD mutant HBV had reduced replicative efficiency in the absence of lamivudine, compared with wild-type HBV (120).

In most cases, YMDD mutations occur in combination with an additional mutation in the B subdomain rtL180M (L528M) that is located in the catalytic site near the YMDD motif in the three-dimensional model of HBV polymerase. The combined L528M/M552I or L528M/M552V mutations have been shown to restore the in vitro viral replication capacity of HBV containing YMDD motif mutations (120). These data reaffirm the growing body of clinical data indicating that patients who develop lamivudine-resistant HBV undergo rebound in serum HBV DNA that returns to level similar to those seen prior to therapy with lamivudine (123). Modeling suggests that mutation of the YMDD motif methionine at position 552 to valine or isoleucine causes steric hindrance between the methyl group on the β-branched side chain of valine/isoleucine and the sulfur atom in the unnatural L-oxathiolane ring of lamivudine (91,121). Besides reducing the strength of binding of lamivudine to the polymerase, steric hindrance results in a different orientation of the inhibitor when bound to the mutant enzyme that reduces the efficiency of incorporation of lamivudine triphosphate into replicating viral DNA. Resistance because of steric hindrance may be a problem common to all L-nucleosides (91). These mutations do not affect binding of the natural substrate dCTP to the same degree because the deoxyribose ring is in the natural D-configuration. Thus, the effect on natural substrates is small compared with the effect on lamivudine, and the polymerase enzyme can preserve a significant, but decreased, level of activity (121).

Adefovir Dipivoxil

Adefovir dipivoxil is an oral prodrug of adefovir (9-[2-phosphonylmethoxyethyl] adenine, PMEA), a phosphonate nucleotide analog of adenosine monophosphate. It is an acyclic nucleoside phosphonate (ANP) compound (124). The antiviral effect of the ANP analogs is the result of a selective interaction of their diphosphate metabolite with the viral DNA polymerase. Based on the structural resemblance to natural deoxynucleoside triphosphates (i.e., dATP in this case), this diphosphate metabolite acts both as a competitive inhibitor and an alternative substrate during the DNA polymerase reaction. PMEA inhibited HBV release from human hepatoma cell lines Hep G2 2.2.15 and HB611 cells (transfected with human HBV) (125). The cells were seeded in 25 cm² tissue culture flasks (Costar, Sigma-Aldrich, St. Louis, MO) at a density of 4 × 10⁴ cells/cm² in Dulbecco’s modified Eagle minimum essential medium (EMEM) supplemented with 2 mmol/L L-glutamine (Flow Laboratories, Rockville, MD), garamycine (40 µg/mL), amphotericin B (2.5 µg/mL), the neomycin analog G418 (360 µg/mL for HepG2 cells; 200 µg/mL for HB611 cells), and 10% FBS. Medium was changed every 3 days. When cells reached confluency at day 6, FBS concentration was reduced to 2%. Cell cultures were maintained in 5% CO₂ atmosphere at 37°C. At day 3, the culture medium was supplemented with various concentrations of PMEA. Cell culture supernatants and cells were harvested at day 12 and subjected to HBV DNA and HBsAg analysis.

The 50% cytotoxic concentration of PMEA was determined in 24-well tissue plates (cell density: 4 × 10⁴ cells/cm²) by inhibition of [³H]methyl-dThd incorporation during 24 hours starting at 3 days after seeding. HBsAg secretion was inhibited by PMEA in a concentration-dependent manner in HB611 cells. Moreover, in congenital duck hepatitis B virus (DHBV)–infected ducklings, PMEA at a dose of 30 mg/kg/day was found to affect a marked decrease in DHBV DNA (122). PMEA has been found to have immunostimulating effects in mice, reflected by the induction of interferon (mainly α/β) and a significant enhancement of natural killer cell activity (126,127). The interferon levels induced by PMEA, given for 5 consecutive days at a dose of 25 mg/kg⁻¹, were comparable with those achieved with the known interferon-inducer poly I:C. The natural killer enhancement by PMEA appeared to be coupled, at least in part, to interferon production. However, cytokines other than interferon may be involved, since prolonged administration of PMEA (5 mg/kg/day for 20 days) resulted in sustained natural killer enhancement in the absence of interferon production.

In two phase III clinical studies, although mutations in the polymerase sites of HBV had been described (rtS119A, rtH133L, rtV214A, rtH234Q) in 4 of 271 patients treated with adefovir dipivoxil for 48 weeks, no adefovir resistance mutations were identified in this large group of hepatitis B patients (128). However, in 1 patient during the second year of treatment with adefovir dipivoxil, the rise of HBV DNA and the exacerbation of liver disease led to the identification of a novel asparagine-to-threonine mutation at residue rt236 in domain D of the HBV polymerase (see Table 11.5). In vitro testing of a laboratory strain encoding the rtN236T mutation and testing of patient-derived virus confirmed that the rtN236T substitution caused a marked reduction in susceptibility to adefovir dipivoxil (129).

Resistance to adefovir dipivoxil is significantly less common than for lamivudine. The emergence of resistance to adefovir dipivoxil appears to be delayed and infrequent. Adefovir is an acyclic nucleotide analog and is smaller than the bulkier oxathiolanes (121). Molecular modeling studies suggest that adefovir can be accommodated more effectively in the more constrained and “crowded” deoxynucleoside triphosphate-binding pocket that carries the rtM204V/I mutations (129). Finally, the adefovir-resistant HBV was sensitive to lamivudine (130).

Entecavir

Entecavir, a cyclopentyl guanosine analog, is a potent inhibitor of HBV DNA polymerase, inhibiting both the priming and elongation steps of viral DNA replication (106,131,132). Entecavir is phosphorylated to its triphosphate, the active compound, by cellular kinases. It is a selective inhibitor of HBV DNA. It has little or no inhibitory effect on the replication of other DNA viruses such as herpes simplex, CMV, and RNA viruses such as HIV. Entecavir is also effective against lamivudine-resistant mutants but less effective than against wild-type HBV (131,133,134). In an in vitro assay using Hep G2 2.2.15 human liver cells, the EC₅₀ for entecavir was 0.00375 µM compared with 0.116 µM for lamivudine. Therefore, entecavir is 30 times more potent than lamivudine in suppressing viral replication. In woodchucks with chronic woodchuck infection, doses of 0.1 mg/kg of entecavir reduced woodchuck hepatic virus (WHV) titers by 7 logs (135). In addition, after 14 months of entecavir therapy, viral core antigen, WHV, and cccDNA were undetectable in liver biopsy samples of nine WHV tested (136). There is also a decrease of the incidence of hepatocellular carcinoma and an increase of survival in treated WHVs as compared with untreated WHVs.

Entecavir has been evaluated in phase I/II clinical studies. The viral dynamics during and after entecavir therapy were studied in a small number of patients with hepatitis B receiving different doses of entecavir ranging from 0.05 to 1.0 mg/day of entecavir. The median effectiveness in blocking viral production was 96%. The median half-life of viral turnover was 16 hours and the median half-life of infected hepatocytes was 257 hours (10.7 days). Rebound of viral replication also followed a biphasic return to baseline levels (137). During short-term therapy, entecavir seems to show stronger antiviral activity than lamivudine, but this assumption should be validated by head-to-head studies. In addition, entecavir shows continuous activity in patients with detectable lamivudine-induced mutant virus (138).

Until now, no entecavir-resistant viral mutants have been described (132). Prolonged therapy as well as prophylactic therapy, for example, in liver-transplant recipients, is feasible and not limited by breakthrough infections. Different ongoing multicenter phase III studies are currently evaluating the efficacy and safety of entecavir in HBeAg-positive and HBeAg-negative patients and in patients resistant to lamivudine. These studies are comparing entecavir with lamivudine for 48 weeks with extended treatment (96 weeks).

Emtricitabine and Tenofovir

Emtricitabine (FTC) is a cytosine nucleoside analog with antiviral activity against both HBV and HIV. It differs from lamivudine in having a fluorine at the 5-position of the nucleic acid. In a pilot study, five different doses of FTC were evaluated (25, 50, 100, 200, and 300 mg daily for 8 weeks) in 49 patients with HBeAg-positive hepatitis B (106,139). At the end of treatment, serum HBV DNA decreased by 2 to 3 logs in patients receiving the higher doses.

Phase III clinical trials are underway to determine the long-term safety and efficacy of FTC. However, the role of FTC in the treatment of hepatitis B may be limited by its structural similarity to lamivudine and hence the potential for cross-resistance and the development of HBV drug-resistant mutants.

Tenofovir is a nucleotide analog approved for the treatment of HIV infection. It has in vitro activity against both wild-type and lamivudine-resistant HBV (140). A report of five patients with HBV and HIV coinfection demonstrated a 4 log₁₀ drop in HBV DNA levels during 24 weeks of treatment (141).

β-L-Nucleosides

The natural nucleosides in the β-L-configuration (β-L-thymidine (L-dT), β-L-2-deoxycytidine (L-dC), and β-L-2-deoxyadenosine [L-dA]) represent a newly discovered class of compounds with potent, selective, and specific activity against hepadnavirus. In vitro studies have shown that these compounds are not active against other viruses such as herpes viruses or HIV, but these compounds have marked effects on HBV replication. It is not yet clear if these compounds are active against lamivudine-resistant HBV mutants (106,142–145).

L-dT is at the most developed stage of clinical investigation and has a remarkably clean preclinical toxicology profile. So far, it does not have mitochondria toxicity and it appears not to be mutagenic. After 1 year of therapy, antiviral activity was significantly greater for L-dT compared with lamivudine, and ALT normalization was greatest for L-dT monotherapy (146). Combination therapy of L-dT and lamivudine was not more efficacious than L-dT alone.

Another promising β-L-nucleoside compound is Val-L-dC. Preliminary results in phase I/II testing indicate substantial antiviral activity with a good safety profile (106).

Combinations of β-L-nucleoside appear to have additive or synergistic effect against HBV. In vitro studies and animal testing showed that there is no evidence of cellular or mitochondrial toxicity. The combination of L-dT and Val-L-dC was analyzed in a woodchuck study. Over a 12-week treatment period, the combination of L-dT and Val-L-dC cleared WHV DNA in all of the five animals tested with no significant side effects noted.

Other Antiviral Drugs

Several compounds have been developed that have a mechanism of HBV inhibition that is unrelated to the viral polymerase. The first of these are the phenylpropenamide derivates, AT-61 and AT-130. King et al. (147) showed that AT-61 did not affect total HBV RNA production or HBV DNA polymerase activity but did significantly reduce the production of encapsidated RNA. Importantly, both AT-61 and AT-130 have identical antiviral activity against wild-type strains as well as a number of different lamivudine-resistant strains of HBV (148). In vitro studies using AT-130 showed significant inhibition of the production of encapsidated HBV RNA but had no effect on total HBV RNA and did not affect core protein or nucleocapsid production, indicating an interference with the encapsidation process itself (149). Steric inhibition or interaction with the host cell chaperone proteins such as Hsp90 may be a possible mechanism. Phenylpropenamides are not water soluble and have very low bioavailability. Their future successful development as antiviral agents will depend on overcoming potential toxicity and medicinal chemistry issues.

A second class of compounds, the heteroaryldihydropyrimidines, are also potent nonnucleoside inhibitors of HBV replication both in vitro and in vivo (150). The heteroaryldihydropyrimidine compounds include the candidate molecule Bay 41–4109 and congeners Bay 38–7690 and Bay 39–5493. Exposure of HBV-infected cells to Bay 41–4109 resulted in increased degradation of core protein through improper formation of viral nucleocapsids. These heteroaryldihydropyrimidine compounds were shown to have efficacy against HBV in the HBV-transgenic mouse model and to possess suitable preclinical pharmacokinetic and toxicologic profile (150). Their novel mechanism of action and highly specific antiviral activity indicates that future clinical studies may be warranted.

A third compound, LY582563, is a 2-amino-6-arylthio-9-phosphonomethoxyethylpurine bis (2,2,2,-trifluoro-ethyl) ester, a novel nucleotide analog derivative of phosphonomethoxyethyl purine. It belongs to a structural class that is similar to adefovir. This compound has excellent antiviral activity against HBV with a good preclinical toxicity profile (151). It is also effective against lamivudine-resistant HBV (152). Its mechanism of action and early clinical development are under investigation.

Outlook: Hepatitis B Therapies

The long-term success of therapy for hepatitis B depends on safe, effective suppression of HBV replication for a long period of time without giving rise to resistant viral strains. As the long-term efficacy of current therapy for hepatitis B is limited, the patient’s age, severity of liver disease, and likely response must be weighed against the potential for adverse events and complications before treatment is initiated. Recommendations for the management of patients with hepatitis B have been recently published by an international panel (153).

In the past several years, the treatment of hepatitis B has been improved. It is now possible to contemplate combination therapy for hepatitis B. The future question is which agents to combine: two or more nucleoside/nucleotide analogs or antiviral agents plus immunomodulatory drugs. A recent report reviews several alternative rescue therapies for chronic hepatitis B (115).

Gene therapy is defined as the introduction of new genetic material into a target cell with a therapeutic benefit to the individual represents a novel approach (154). Several genetic antiviral strategies including ribozymes, antisense oligonucleotides, interfering peptides or proteins, and therapeutic DNA vaccine have been explored for the molecular therapy of hepatitis B (154,155). A novel molecular strategy that holds promise is the use of small interfering RNAs (siRNA). RNA interference is a cellular process of sequence-specific gene silencing in which small duplexes of RNA target a homologous sequence for cleavage by cellular ribonucleases (156). The introduction of approximately 22-nt siRNAs into mammalian cells can lead to specific silencing of cellular mRNAs without induction of the nonspecific interferon responses that are activated by longer RNA duplexes. Posttranscriptional gene silencing mediates resistance to both endogenous, parasitic, and exogenous pathogenic nucleic acids and can regulate the expression of protein-coding gene (156). This approach has been successfully applied to HBV (157). The siRNA molecules dramatically reduced virus-specific protein expression and RNA synthesis, and these antiviral effects were independent of interferon. Although this approach holds great promise, issues such as gene delivery, stability, toxicity, resistance, and safety need to be resolved.

Hepatitis C virus (HCV) is a major pathogen, harbored by over 250 million people worldwide, with a 70% risk of chronic infection for contaminated individuals (158). Chronic hepatitis may lead not only to limited histologic lesions in 20% of patients but also to severe fibrosis and cirrhosis. Complications arising from chronic HCV infection include the development of cirrhosis, end-stage liver disease, and hepatocellular carcinoma. Among the latter, at least 30% will develop hepatocellular carcinoma in the following 20 years (158,159). Accordingly, complications arising from chronic HCV are the leading cause of death from liver disease and the most common indication for liver transplantation throughout the world. Although recent progress has been made in the knowledge of the HCV life cycle, thanks to new experimental tools such as selectable subgenomic replicons, we are still limited in our understanding by the lack of an efficient cell culture system and of an easy-to-use animal model (160). Consequently, the identification of new therapeutic targets is particularly difficult. Interferon-α (IFN-α) and ribavirin have been the mainstay of available therapies for chronic HCV infection for years. Therapeutic results may appear disappointing as, at best, 50% of the treated patients will have undetectable HCV viremia when these drugs are used in combination therapy (161).

Therapeutic improvements resulted from the progression of using standard IFN-α monotherapy to pegylated interferons coupled with ribavirin when measured in the rates of sustained virologic response (SVR). Despite these improvements, the rates of cure in genotype 1 infection remained around 40% to 50% for treatment-naive individuals with poorer outcomes for African Americans, individuals with cirrhosis, prior treatment failure, and HIV coinfection (162–165).

Ongoing research in the elucidation of the HCV life cycle and crystal structures of several viral proteins has broadened the development of drug targeted at specific points in the HCV life cycle (166–169) (Table 11.8). Specifically, it has been the nonstructural (NS) protein required for ongoing viral replication and spread that have been the primary targets of drug development. Direct-acting antiviral agents (DAAs), which include NS3/4A PIs, replication complex inhibitor (NS5A), and NS5B polymerase inhibitor, have been developed (see Table 11.8). Additionally, agents that target host protein involved in viral replication have been studied

Biology of Hepatitis C Virus

HCV is an enveloped virus containing a single-strand, positive-sense RNA (170). It belongs to the Hepacivirus genus of the Flaviviridae family, also containing two other genera: pestivirus and flavivirus. The most closely related virus is GBC-B, a hepatotropic virus infecting tamarins (171). The HCV genome contains approximately 9,600 nt with one ORF, encoding for a single polyprotein.

Genetic Variability of Hepatitis C Virus

Despite this apparently basic genetic structure, the polyprotein may vary, depending on the genotype. In fact, HCV RNA shows wide genetic variability as the estimated rate of nucleotide change is around 10³substitutions per site per year (172), although this rate of nucleotide change depends on the considered genomic region. For example, the polyprotein is flanked by two nontranslated regions that are highly conserved and in contrast, some hot spots of mutations have been recognized, particularly in the E2 envelope protein, which contains two hypervariable sequences (173). It is noteworthy that this genetic variability complicates study designs of HCV pathogenesis since a single viral protein may lead to different in vitro reactivity with host proteins, depending on the amino acid sequence. Thus, over 90 genotypes have been screened around the world, and six main HCV types are now distinguished, according to the proposed classification (174). Prevalence of genotypes may vary around the world; for example, genotype 1b represents more than 45% of the isolates in Europe and only 17% in the United States (175). This genetic variability is obvious within the host, representing quasispecies (176). These facts are, at least in part, related to the high viral turnover, since the estimated viral production is 10¹² per day (174,177).

Since infectious virions are permanently selected through their interactions with the host, this may partly explain immune failure to eradicate HCV and resistance to antiviral drugs (160,178). However, the identification of some highly conserved amino acid sequences among both structural and nonstructural proteins suggest essential functions for viral life cycle and could constitute the target of new antiviral drugs.

Viral Life Cycle

The polyprotein is processed by viral and cellular proteases in 10 structural and nonstructural proteins. Structural proteins include the core (i.e., the viral nucleocapsid) and the envelope proteins El and E2. The NS proteins are NS2 to NS5B, required for viral replication. These two groups of proteins are separated by the short membrane peptide p7 (160,179). Recently, the p7 function appeared to act as an ion channel (180). This assemblage results in a complete virion exhibiting a diameter of 30 to 50 nm (181). However, particles isolated from plasma may vary as a result of complex formation with very-low-density lipoproteins or immunoglobulins (182).

The main cellular target for HCV is the hepatocyte (177). But extrahepatic sites allowing viral replication have been identified, such as PBMCs (essentially B cells and monocytes), lymph nodes, and possibly biliary cells (177,183,184). More than one cellular receptor should exist for HCV: CD81 was the first one to be discovered (185), but experimental animal models showed that it did not allow viral replication by itself (186). Accordingly, very-low-density lipoprotein receptors and glycosaminoglycans are potential coreceptors for HCV (187).

The life cycle of HCV is more hypothetical beyond the stage of target cell infection because our progress in this field is hampered by the absence of a relevant model for HCV entry, replication, and release. Clearly, El and E2 are involved in membrane fusion operating between the virus and the endoplasmic reticulum (188). For this purpose, these two viral products interact through heterodimerization, allowing endoplasmic reticulum retention (189). Thus, viral RNA translation may proceed, beginning by the internal ribosome entry site binding to ribosome (190,191). The positive-stranded viral RNA serves as a template for the production of RNA negative strands, which allow the synthesis of viral genome, and thus virions, by interactions with copies of the core protein (190). As an example, the core protein produced in bacteria allows the formation of nucleocapsids when incubated with RNA molecules (192). However, NS proteins such as NS3 helicase and NS5B RNA-dependent RNA-polymerase obviously play a major role in this viral life cycle, but our understanding of their molecular interactions needs to be improved (160).

Experimental Systems for Antiviral Drug Study

Attempts to explore and understand the antiviral activity of IFN-α and ribavirin have been studied (193–195). For an in-depth understanding of the mechanism of action, several investigators consider that an experimental system is necessary. Toward this end, HCV protein functions and their interaction with the crystal structure of the RNA-dependent RNA polymerase has been determined (196). Dellamonica et al. (196) described several culture model systems using cellular or animal models. These include fetal liver cells, hepatoma cell lines, immortalized T or B cell lines, and a simian model that mirrors the natural history of HCV infection (197–204).

All this information is required to fully understand the current procedure to study the antiviral effects of licensed drugs. However, the mechanisms accounting for the antiviral activity of both IFN-α and ribavirin are essentially unidentified (158,161,193,194). IFN-α is thought to induce an antiviral cellular state and may upregulate the destruction of infected cells. Ribavirin is a synthetic guanosine analog mostly used for the treatment of respiratory syncytial virus infection. The inhibition of HCV is hypothetical and it seems that most of its antiviral effect is indirect through synergic interaction with IFN-α, upregulating antiviral immune responses (195).

Evaluating Antiviral Therapy

For HCV infection, viral load monitoring to measure HCV RNA levels in serum or plasma has played a critical role in assessing the efficacy of antiviral therapy. Achievement of SVR, defined as undetectable HCV RNA in serum or plasma, 12 to 24 weeks after completion of therapy is a primary objective of anti-HCV treatment and has been interpreted as clinical “cure” (205–207). Undetectable levels are typically interpreted as 15 IU/mL of HCV RNA. A report of “undetectable” HCV RNA does not necessarily mean the absence of virus but rather that the assay may be unable to detect HCV RNA because the level is below the lower limit of quantitation (LLOQ) and outside of the linear range where the manufacturer (or laboratory) can attest to a highly reproducible and accurate measurement.

An additional use of viral load monitoring of HCV RNA levels during treatment is to guide duration of therapy, known as response-guided therapy (RGT). RGT has become a necessary and complementary component of patient management protocols because it represents a personalized approach that can optimize treatment, safety, and outcomes (208,209).

Outlook: Hepatitis C

Despite many unresolved aspects of HCV infection, considerable efforts have been made since the definitive identification of the virus. A better understanding of HCV epidemiology will allow improved prevention; meanwhile, detection of infected patients is easy to achieve early in the natural history of their disease. Therefore, treatment combinations may be prescribed in cases of mild liver disease, increasing chances of cure for patients. Moreover, current tools for the study of the HCV life cycle should provide the opportunity to develop new antiviral drugs.

There appears to be a dynamic drug pipeline for HCV treatment where the goal will be an HCV regimen free of interferon with orally administered treatment. The landmark introduction in 2011 of the initial two DAAs telaprevir and boceprevir for HCV genotype 1 infections may have been the prelude to this goal. However, it was disappointing that results from clinical trials preclude their administration as monotherapy. Both compounds have a low barrier to resistance and require concomitant administration of pegylated interferon and ribavirin to prevent the emergence of resistant mutants.

Human cytomegalovirus (HCMV), a member of the herpesvirus family, is a ubiquitous agent that commonly infects human beings. As with other herpesviruses, primary infection is followed by latent infection. Recurrent infections are most often caused by reactivation of latent virus but reinfection also occurs. HCMV infections are generally mild or asymptomatic in immunocompetent adults, but HCMV is a major cause of defects in neonates and is a major pathogen in immunosuppressed individuals, including recipients of bone marrow and solid-organ transplants and patients with AIDS. In particular, HCMV was the most common cause of sight- and life-threatening opportunistic infection in patients with AIDS prior to the availability of highly active ARV therapy.

The cornerstone of antiviral therapy is ganciclovir, which was the first compound licensed in the United States specifically for treatment of CMV infections. Ganciclovir is a synthetic acyclic nucleoside analog, structurally similar to guanine. Its structure is similar to acyclovir and, like acyclovir, requires phosphorylation to achieve antiviral activity. Following phosphorylation by the viral protein pUL97, cellular enzymes phosphorylate the monophosphate form to di- and triphosphate metabolites. Subsequently, the ganciclovir-triphosphate metabolite then exerts its antiviral effect on CMV-infected cells.

Other compounds currently approved for the treatment of severe HCMV infections include the deoxyguanosine analog ganciclovir and its prodrug valganciclovir, the ANP cidofovir, and the pyrophosphate analog foscarnet. All of these compounds are inhibitors of the viral DNA polymerase encoded by gene UL54.

As cidofovir already contains a phosphate-mimetic group, it needs only two phosphorylation steps to reach the active stage. These phosphorylations are performed by host cellular enzymes. Thus, cidofovir does not depend on the virus-induced kinase to exert its antiviral action. Cidofovir diphosphate interacts as competitive inhibitor with the normal substrate (deoxycytosine triphosphate) for the viral polymerase. Two consecutive incorporations at the 3′ end of the DNA chain are required to efficiently shut off DNA elongation. Incorporation of one cidofovir diphosphate molecule causes a marked decrease in the rate of DNA elongation. Foscarnet does not require intracellular activation to exert its antiviral activity. It is not incorporated into the growing DNA and it reversibly blocks the pyrophosphate binding site of the viral DNA polymerase and inhibits the cleavage of pyrophosphate from deoxynucleoside triphosphates.

Patients receiving long-term suppressive anti-HCMV therapy may develop antiviral resistance. In AIDS patients before the era of highly active ARV therapy, epidemiologic studies have shown that the frequency of resistance increased with the duration of therapy and reached 27% of patients after 9 months of ganciclovir (210), 13% of patients after 1 year of valganciclovir (211), and 37% of patients after 1 year of foscarnet (212). In solid organ recipients, 7% of the HCMV infection were resistant to ganciclovir after a median delay of 10 months (213).

Methods used in the laboratory to determine the susceptibility of HCMV strains to antiviral drugs are classified as phenotypic or genotypic. Phenotypic methods aim to determine the concentration of an antiviral agent that inhibits the virus in culture. Genotypic assays are designed to detect mutations known to confer antiviral resistance in the genome of the viruses being studied.

Phenotypic Antiviral Assays

A variety of phenotypic assays have been used in different studies to determine the antiviral susceptibilities of HCMV strains. Each of these methods measures the inhibition of HCMV growth in tissue cultures in the presence of serial dilutions of antiviral drugs. The 50% inhibitory concentration (IC₅₀) is defined as the concentration of antiviral agent resulting in a 50% reduction in viral growth. The methods usually performed include the PRAs that measure the inhibition of replication of infectious virus, the assays based on DNA hybridization that measure the inhibition of viral DNA synthesis, and enzyme-linked immunosorbent and flow cytometry–based assays that measure production of one or more viral proteins. The IC₅₀ values determined with these different phenotypic assays depend on the nature of the replication marker chosen.

Clinical Specimens to Study

HCMV can be recovered from various clinical specimens such as peripheral blood leukocytes, bronchoalveolar liquid, cerebrospinal fluid, organ biopsy, urine samples, and vitreous fluid. Strains recovered from blood are usually chosen to be tested because viremia reflects blood dissemination and active infection. However, resistance profiles of a blood strain may differ from those of strains recovered from other body compartments and directly responsible for the disease because antiviral drug selection pressure may differ from one compartment to another. In addition, patients may shed multiple strains of HCMV either concurrently or sequentially (214).

Constitution of Viral Stocks

All the assay methods require the constitution of viral stocks obtained after sequential passaging of the viral isolate. Viral stocks are constituted by either extracellular virus recovered from culture supernatants or intracellular virus present in infected cells. Constitution of extracellular virus stocks requires at least 8 to 10 passages and is therefore time-consuming. Moreover, some isolates remain cell associated and no extracellular virions are produced. On the contrary, constitution of a stock of infected cells may be achieved after only one to three passages; however, it may require more passages (215).

Tissue culture supernatants containing extracellular virus are clarified by centrifugation at 1,000 rpm for 10 minutes and divided into aliquots and kept frozen at −80°C until use. Titers of extracellular virus stocks are determined from a thawed aliquot using a plaque assay or a rapid immunocytochemistry assay. Infected fibroblast monolayers are trypsinized; cells are then counted and used for the phenotypic assay.

Plaque Reduction Assay

PRA has been considered as the “gold standard” for antiviral phenotypic testing. A standardized assay has been proposed by Landry et al. (216).

Human fibroblasts grown as just confluent monolayers in 24-well plates are inoculated with a standardized inoculum of a stock virus (50 to 100 plaque-forming units per well). After adsorption for 90 minutes, the medium is aspirated and the wells are overlaid with a 0.4% agarose medium containing serial dilutions of the drug to be tested. Concentrations usually tested are between 1 and 50 µM for ganciclovir, 50 and 800 µM for foscarnet, and 0.1 and 20 µM for cidofovir. Each drug concentration is tested at least in triplicate, as well as controls without drug. The cultures are incubated for 7 days at 37°C. Monolayers are fixed in 10% formalin in phosphate-buffered saline (PBS) and stained with 0.8% crystal violet in 50% ethanol. The IC₅₀ of the antiviral agent for the isolate is defined as the concentration causing a 50% reduction in the number of plaques produced as compared with controls. Alternatively, a medium without agarose can be used. The plates are then incubated for 4 days and the foci are revealed by immunoperoxidase staining using monoclonal antibody E-13 (214).

DNA Hybridization Assay

Cell cultures are inoculated with a standardized amount of virus and incubated in the presence of different concentrations of antiviral agent until control wells show 60% to 80% cytopathic effect. Then the cells are lysed and whole DNA is extracted and hybridized to a radiolabelled HCMV probe. Radioactivity is counted. The IC₅₀s of the antiviral agents are the concentrations that reduce by 50% the DNA hybridization values compared with the hybridization values of controls (217).

Detection of Human Cytomegalovirus Antigens

Viral production is measured by using immunofluorescence-, immunoperoxidase-, or flow cytometry–based methods for detection and quantification of cells expressing HCMV antigens or by using enzyme-linked immunosorbent assay–based methods for quantification of HCMV antigens produced in cells (218,219).

Limitations of Phenotypic Assays

Phenotypic assays are limited by the excessive time required to complete the assay (1 to 2 months). To reduce the turnaround time involved in PRAs, a modified assay has been performed directly in primary cultures of clinical specimens such as urine, amniotic fluid, and bronchoalveolar lavage samples (220). Blood leukocytes from patients with documented HCMV viremia were also used as inoculum in a PRA (221). These methods provided results within 4 to 6 days. Their major limitation is that the virus titers of the clinical specimens are often too low to allow them to be used.

Resistant strains in a mixture may not be detected because of strain selection that occurs during passaging. Therefore, the virus stock studied may be not representative of the original population of virus. If antiviral drug is added to the cell culture medium in order to favor growth of mutant virus, de novo resistant strains may be selected in cell culture (222).

Interpretation

Dose-inhibition curves are constructed to determine the IC₅₀ values of the antiviral agents of the studied strains. The type of cell culture used, the size of the viral inoculum, the method used, and the laboratory performing the test are factors that affect the results. Therefore, there is a significant variability among methods and laboratories. When feasible, a baseline isolate from the same patient should be tested in parallel with the isolate of interest. However, baseline isolates are often missing. Reference susceptible and resistant strains must be included in each susceptibility assay. Antiviral drug IC₅₀ values for the isolate to test are compared with the results obtained for the reference strains. A comparison with the results obtained for a panel of susceptible strains tested in the laboratory can assist in the interpretation.

The IC₅₀ cutoff values proposed to define resistance to antiviral compounds are as follows: more than 12 µM for ganciclovir, more than 400 µM for foscarnet, and more than 2 or 4 µM for cidofovir. HCMV strains for which ganciclovir IC₅₀ is more than 6 µM and less than 12 µM are considered to have decreased susceptibility to ganciclovir (215,216,223). However, cutoff values are to be determined in each laboratory.

Genotypic Antiviral Assays

Ganciclovir resistance results mostly from changes in the UL97 phosphotransferase responsible for the primophosphorylation of ganciclovir (Table 11.9) (224,225). Mutations in the UL54 DNA polymerase gene appear after prolonged ganciclovir therapy (226). They contribute to a high level of resistance to ganciclovir and induce cross-resistance to cidofovir. All foscarnet and cidofovir resistance mutations that are currently known map to the UL54gene (Table 11.10) (223,227–231).

Viral DNA Extraction

Viral DNA is extracted from cultures in 25-cm² flasks using the procedure described by Hirt (232). Briefly, infected cell monolayers exhibiting at least 50% cytopathic effect are washed twice in PBS and then 0.4 mL of Hirt solution (100 mM Tris-HCl, pH 7.5, 10 mM EDTA, 0.5% SDS) is added. After incubation at room temperature for 20 minutes, the lysate is transferred to a microtube and 0.1 mL of 5 M NaCl solution is added. The mixture is incubated for 12 hours at 4°C and centrifuged at 15,000 rpm for 5 minutes. The resulting supernatant is digested with proteinase K (1 mg/mL) at 56°C for 1 hour and then extracted with phenol chloroform and precipitated with ethanol. By this procedure, a purified DNA preparation enriched in viral DNA is obtained. Alternatively, infected monolayers are trypsinized, washed twice in PBS, and centrifuged. The cell pellet is submitted to lysis in a buffer containing 10 mM Tris-HCl, pH 8.0, 50 mM KCl, 2 mM MgCl₂, 0.9% Nonidet P-40, and 100 µg/mL of proteinase K for 1 hour at 56°C (215). The mixture is centrifuged for 5 minutes at 15,000 rpm and the supernatant containing DNA is collected. DNA is extracted from clinical samples using commercially available purification columns according to the manufacturer’s instructions.

Rapid Genotypic Methods

Detection of Ganciclovir Resistance-Related Mutations in Gene UL97.Rapid screening methods have been designed to detect the mutations at codons 460, 594, 591, 592, 594, 595 (the most frequently observed in clinical isolates), and at codon 520 (233–236). They are based on restriction enzyme analysis of three selected PCR products amplified from HCMV-infected cultures or from clinical samples (Tables 11.11 and 11.12). The PCR reaction mixtures include 5% dimethyl sulfoxide, deoxynucleosides triphosphate at 200 µM each, primers at 1µM each, 2.5 units of Taq polymerase (Roche Diagnostics, Basel, Switzerland), and 500 ng of DNA. The reactions are cycled 40 times as follows: 95°C for 1 minute, 55°C for 45 seconds, and 72°C for 45 seconds.

The restriction enzymes Nla III, Alu I, Hha I, Mse I, Taq I, and Hae III are used to digest the PCR products (Table 11.13) (233,235,236). The digests are analyzed on a 15% polyacrylamide gel and stained with ethidium bromide.

The diagnostic screening assays based on restriction enzyme analysis of selected PCR products allow the detection of the most frequent UL97 mutations involved in resistance to ganciclovir and can identify minority mutants if they reach 10% of the viral population. The disadvantage of these rapid and simple tests is that they miss specific mutations that do not result in a change in the restriction enzyme pattern and previously unmapped mutations.

Detection of Resistance-Related Mutations in Gene UL54.Restriction enzyme analysis of PCR fragments have been used to detect the ganciclovir and cidofovir resistance change L501F and the ganciclovir and foscarnet resistance change A809V in the DNA polymerase (228,237).

Sequence Analysis Methods

Sequencing is the nucleotide sequence analysis that is proven to be the most accurate method for genotypic resistance determination of HCMV strains. However, no standardized assay is available to routinely achieve the analysis of the UL97 and UL54 genes. The complete genes or the gene regions involved in resistance are amplified by PCR and then PCR products are directly sequenced (Tables 11.14 to 11.16) (215,238,239). Amplification of UL97, either with the primer set CPL97-F and CPL97-R or the primer set HLF97-F and HLF97-R, is performed using the GeneAmp XL PCR kit (Applied Biosystems, Foster City, CA). The reactions are cycled 30 times as follows: 94°C for 1 minute and 60°C for 10 minutes. The entire gene UL54 is amplified using the GeneAmp XL PCR kit (Applied Biosystems, Foster City, CA). The UL54 region of interest is amplified using high-fidelity polymerase. PCR products are purified using QIAquick PCR (Qiagen, Valencia, CA) purification kits and are then submitted to sequence reaction. When analysis of sequences are to be performed on an ABI automated DNA sequencer, the purified templates (10 or 20 ng) are sequenced using ABI Prism BigDye Terminator version 3.0 (Applied Biosystems, Foster City, CA) ready reaction cycle sequencing kit according to the manufacturer’s instructions. Depending on the sequencing method chosen, minority mutants in a mixture can be detected if they reach 20% to 40% of the viral population. Sequences of UL97 and UL54 are aligned with the AD169 strain reference sequence (EMBL accession no. X17403) using Align Plus, version 4.0 (Scientific and Educational Software, Durham, NC).

The interpretation of genetic assays requires the distinction of resistance-associated mutations to natural interstrain variation. Some of the mutations observed in resistant strains have been validated as resistance markers by a process of marker transfer, but others have not. To date, resistance mutations in UL97 have been found at one of three sites. At two of those, there are point mutations in a single codon (460 and 520), and at the third site, there are point mutations or deletions within the codon range 590 to 607. On the contrary, resistance-associated mutations in UL54 are widely dispersed across the coding sequence. The interstrain sequence homology is above 98% and 99% for UL54 and UL97, respectively (215,238,239). UL54 sequences determined in HCMV isolates sensitive to antiviral drugs have been deposited in GenBank under accession numbers AF133589 through AF133628 and under accession numbers AY422355 through AY422377 (238,239). UL97 sequences of ganciclovir-sensitive strains have been deposited under accession numbers AF34548 through AF345573 (215).

Outlook: Human Cytomegalovirus Assays

Phenotypic assays are still important to corroborate genotypic results. However, the time required to perform the assays is too long to provide useful therapeutic information. Results obtained by both types of susceptibility testing are usually concordant. However, it should be noted that viruses harboring mutations responsible for resistance can be classified as susceptible in phenotypic assays. Because some UL97or UL54 mutations are associated with a slow growth of HCMV in culture, it has been proposed that the presence of these mutations could introduce bias in the interpretation of phenotypic assays. As patients can be infected by multiple HCMV strains, cell culture–based assays can select a different virus population from that selected by PCR-based assays.

In the specialized area of CMV viral load testing for organ transplant donors and recipients, a notable development was recently unveiled, which is expected to transform clinical practice. WHO revised the solid organ transplant guidelines, which require reporting viral load results using an international standard (240). The revised guidelines clearly demonstrate that CMV monitoring has become part of the PCR era with quantitative PCR viral load testing as the standard of care for monitoring patients. Toward this end, FDA recently approved the first CMV viral load assay with traceability to the WHO international CMV standard. These developments have garnered significant optimism in this field because it will make possible universal clinical recommendations for CMV patient management and will positively impact outcomes.

Influenza continues to be an important public health concern in the 21st century. This infection is responsible for annual epidemics with significant mortality and morbidity (241,242). During recent outbreaks, pandemic strain of influenza A emerged from the animal reservoir, leading to devastating influenza with high attack rates, as during the H1N1 “swine flu” pandemic in 2009 to 2010 (243,244). Pandemics were observed three times during the 20th century (245,246) and most recently in the 21st century during 2009 to 2010 (242,244). Influenza is highly contagious, with person-to-person contagion spread easily via aerosol droplets that infect epithelial cells of the respiratory tract (247–249).

Human and animal influenza viruses belong to the family Orthomyxoviridae. There are three genera corresponding to influenza virus types A, B, and C (250,251). Influenza viruses are divided into subtypes based on major antigenic specificities of their surface glycoproteins HA and NA. To date, 16 different HAs and 9 different NAs have been described. The name of the influenza virus each season reflects the HA and NA segments present in the strain following gene reassortment. In 2013, influenza A (H1N1) and A (H3N2) represented the typical seasonal influenza A virus subtypes. Influenza A (H1N1) is the same strain that caused the 2009 influenza pandemic and in 2013 is circulating as seasonal influenza (242). Two influenza B viruses also cause seasonal influenza. Influenza B strains are named after the communities in which they first appeared, Victoria and Yamagata strains (242). Type C influenza causes sporadic and only minor outbreaks, although occasionally may cause more severe illness in children (252). In recent years, sporadic transmission of influenza viruses from animals to humans has occurred. In 2011, swine influenza virus subtype A (H3N2) was detected in people; the label “variant” influenza (vH3N2) was used to distinguish these viruses from human viruses of the same strain (242,244).

Among influenza viruses, influenza A is most notable. In recent years, influenza B caused less than 20% of clinical cases of influenza. Once infection occurred, however, it was able to produce significant morbidity (244). Both viruses together cause annual outbreaks and epidemics during cold winter seasons in temperate climates. These organisms are also responsible for epidemic illness throughout the world (245). During some recent warmer winters attributed to climate change, fewer people became infected with influenza and hence had waning immunity. These seasons were followed by more severe epidemics during subsequent years (253).

Influenza is associated with nonspecific signs and symptoms (250). The definition for influenza-like illness is the triad linking fever plus one respiratory and one general sign. Briefly, following a short incubation period of 24 to 48 hours, the disease is usually characterized by the sudden onset of high fever with chills, headache, dry cough, myalgia, asthenia, diarrhea, and nasal congestion with or without mild rhinorrhea (249,250,254). This disease is observed in all age groups. In some cases, bacterial superinfection can be observed, leading to severe pulmonary infection and increased morbidity and mortality (255,256). In classic influenza illness, spontaneous recovery is observed after 5 to 7 days of infection, depending on the patient’s age and the immune status. In the recent outbreaks, populations more susceptible to complications and mortality included the elderly, young children, pregnant women, those with chronic respiratory diseases, the morbidly obese, as well as immunocompromised individuals (249,257,258). People older than 65 years of age accounted for 90% of seasonal influenza–associated deaths, even though these individuals made up only about 15% of the population (259). In general, the disease lasts 7 to 8 days in young children and can last several months in immunocompromised patients.

Prevention strategies for high-risk populations have been implemented, given the seriousness of this disease. Since 2010, annual influenza vaccination has been recommended for everyone older than 6 months of age (249,260). Most hospitals and health care institutions mandate vaccination of all health care workers as well (260). Because influenza C has minimal population disease risk as compared to influenza A and B, only the A and B strains are included in seasonal influenza vaccines (242). Many institutions prohibit health care workers from coming to work when they have respiratory symptoms and fever during times of epidemic. Wearing of face masks, compliance with handwashing protocols, and respiratory droplet isolation are all important components of influenza prevention programs (244). Public health considerations include administration of antiviral agents to nonvaccinated patients within 48 hours of influenza exposure, especially to those who are at high risk of influenza complications due to premorbid conditions or who are severely immunosuppressed (244).

Influenza Virus Structure

Influenza A and B viruses share common features (250). They are pleomorphic, 80- to 120-nm virus particles. Influenza A, B, and C viruses can be differentiated by their viral-specific nucleoprotein and matrix proteins (252). The viral membrane is covered with a large quantity of evenly spaced glycoproteins: HA and NA, with four times more HA than NA molecules. Influenza A viral RNA has eight segments of single-stranded negative-sense RNA and many copies of nucleoprotein (NP), along with the polymerase complex. Six of the eight segments of RNA (see Table 11.17 for a list of names and functions) each produce a single viral protein, and two segments (7 and 8) produce two proteins each. The seventh segment encodes for the matrix 2 protein (M2) and the matrix 1 protein (M1). The ribonucleoprotein (RNP) and the lipid envelope are connected by the M1 protein. Influenza B and C also contain polymerase proteins and NP. However, influenza B has only four membrane proteins (HA, NA, influenza B matrix 2 protein, and NB, a membrane protein specific to influenza B); influenza C contains only two proteins, influenza C matrix 2 (CM2) and the HA-esterase-fusion (HEF) protein, which functions as HA and NA combined (251,261).

These are all enveloped viruses. Influenza A viral envelope is made of a lipid bilayer containing three transmembrane proteins: HA (which makes up 80% of the membrane protein), NA (17% of the membrane protein), and M2 protein (making up a minor component of the membrane, with up to only 20 molecules per virus) (251). HA and NA are exclusively within the viral envelope membrane and M2 forms tetramers through the membrane. M1 protein is under the lipid bilayer (251).

Protein Function

Matrix 1 Protein

M1 protein is responsible for the stability of the membrane and interacts with the virus RNA segments. M1 lies just beneath the lipid membrane and attaches to the viral ribonucleoproteins (vRNPs). vRNPs are composed of single-stranded, negative-sense RNAs wrapped around nonstructural protein 1 (NSP1) and nonstructural protein 2 (NSP2; also called nuclear export protein or NEP). Three viral polymerase proteins make up the rest of this complex (250,251,262).

Influenza RNA is surrounded by NP. The complexed RNP is transcribed by the viral RNA polymerases to make mRNAs leading to viral protein synthesis. The NP-RNA complex sits in the lipid bilayer membrane, together with the HA and NA glycoproteins and the proton channel M2 protein. NS1 is not packaged into new virus particles but has important interactions with the host innate immune system and plays a role in viral growth and pathogenicity. NS2/NEP controls interaction between viral RNA and vRNP complexes to assist in the export new viral particles from the host cell (259).

Matrix 2 Protein

Matrix 2 (M2) channel protein found within the viral envelope is identified in influenza A viruses only (250,262). To be active, M2 molecules are organized as disulfide-linked tetramers, located as integral membrane proteins, with their N terminus expressed at the virus surface. They are encoded by a spliced mRNA from genome RNA segment 7 of the virus (250). The active part of M2 in the ion channel is located in the 19 residue transmembrane domain (251). During acidification of the endosome, these tetramers form an important channel for hydrogen ions between the interior of the virus and its environment, providing a low pH within the virus, leading to uncoating. Acidification via the M2 channel also allows the M1 protein to dissociate from the RNP-NP complex, permitting viral RNP to enter the nucleus of the cell and begin viral replication.

Hemagglutinin

HA is one of two major virus surface proteins and is encoded from genome RNA segment 4 in influenza A and B viruses (250). HA is a complex protein with key functions in initiating the infection process through binding to sialic acid receptors (263,264). It is synthetized as a precursor protein, HA0, which has to undergo a proteolytic cleavage to become active. Cleavage generates two proteins (HA1 and HA2) bonded by two disulfide bonds (Fig. 11.4). HA1 is of globular form and contains the major antigenic sites as well as the receptor binding site (RBS), which is highly conserved between types and subtypes (265,266). On host cells, specificity of viral binding to glycoproteins are determined by two species-restricted linkages: α(2,3) (avian recognition) and α(2,6) (human recognition). Viruses from swine recognize and bind to either. HA molecule attachment to cell surface sialic acid receptors are strongly dependent on these binding sites (251). HA2 is a long, fibrous stem that forms triple-stranded, coiled-coil protein that contains the transmembrane domain and the fusion peptide; this fusion peptide is available after HA0 cleavage (267). At the virus surface, the HA glycoproteins are trimeric and very abundant. When submitted to low pH, HA undergoes conformational changes that allow HA-mediated membrane fusion (263). Subsequently, endosomes form, allowing the M2 ion channel to open which acidifies the viral core, releasing vRNP to enter host cell cytoplasm (251). vRNP must then enter the host nucleus in order for viral transcription and replication to take place, culminating in development of host infection.

Neuraminidase

NA is the second major surface glycoprotein (250,268). It is encoded from genome RNA segment 6 in influenza A and B viruses. It is organized into two domains, the stalk and the head. The head of the NA contains the active site of the enzyme (Fig. 11.5). These proteins are organized as homotetramers at the virus surface (250). Among influenza viruses, the structure of the catalytic site of NA is highly conserved. This protein is a sialidase. The role of NA is to cleave sialic acids from adjacent glycoproteins, preventing HA from aggregating with the cell surface sialic acids. Effectively, NA facilitates release of newly produced viruses from the host cell surface after budding. Without this process, viral particles would not be released from the plasma membrane (251). Efficient interplay of HA and NA activities is necessary for optimal viral infection. After release, NA enables the virus to find target cells despite the large amount of protein covered with sialic acids in nasal mucus (268). Compared with influenza A, on the surface of influenza B, NA is present in very low quantity (252).

BM2 Protein

Influenza B matrix 2 protein (BM2) is found in influenza B only (250,252). It is a tetramer protein similar to the M2 influenza A matrix protein. The RNA segment 7 of the influenza B virus codes for this protein, similar to the M2 protein of influenza A. BM2 protein produces a pH-sensitive proton channel that is mandatory for successful virus replication. These proteins appear to have dual functions of transmembrane proton conductance as well as recruitment of other proteins during virus replication (252).

Although M2 and BM2 are structurally and functionally similar, there is limited homology between their amino acid sequence and only in the transmembrane portion of the protein. These differences in function are highlighted by the observation that the M2 channel but not the BM2 channel is inhibited by the antiviral drug amantadine (261).

Viral Infection and Pathogenic Mechanisms

The first step in cellular infection by influenza viruses is the attachment of the virus to its cellular receptor (246,250,251,263). These receptors are identical for influenza A and B viruses. This attachment is performed by interaction between the RBS located at the end of the HA and the sialic acids of the target cell. Virus attachment depends on the recognition of the specific sialic acid by the RBS (251,263). There are sialic acids with different structures; the one recognized by human influenza viruses is linked to galactose by the α(2–3) linkage. This attachment mediates endocytosis. After endocytosis, the pH in the endosome is lowered, leading to HA conformational change and subsequent membrane fusion. The fusion peptide of several HA trimers intercalate into lipid bilayers to be efficient (251,252,263). Concomitantly, the M2 ion channel conducts H⁺ ions inside the viral particle to allow uncoating and subsequent release of vRNPs in the cytoplasm of the infected cell and its transport to the nucleus. vRNAs are then transcribed into mRNAs and replicated.

Viral mRNA synthesis in the nucleus requires initiation by host cell primers. Replication of virion RNA occurs in two steps: the synthesis of the RNA template (full-length copies) and the copy of these templates into vRNAs. Viral mRNA is translated into viral proteins. HA, NA, and M2 are integral membrane proteins and are synthesized in the Golgi apparatus, while M1, NP, PA, PB1, PB2, NS1, and NEP are not. During synthesis in the Golgi apparatus, M2 protein acts as an ion channel to prevent HA molecules from undergoing early, irreversible, pH-induced conformational changes. Synthesis of HA in the endoplasmic reticulum is a stepwise conformational maturation of the protein with independent folding of specific domains in the HA monomer. This is followed by trimerization of uncleaved HA0 and the completion of folding to a pH-neutral form of HA. Once properly assembled and folded, it is transported to the Golgi apparatus where oligosaccharide chains are added and HA0 cleavage to HA1 and HA2 is performed. NA and M2 are eventually are transported to the cell surface. These integral proteins are expressed at the cell surface, preferentially in lipid raft microdomains, and can initiate viral budding (251).

The budding process starts by protein–protein interaction between intracellular domains of the NA, M2, and HA and M1 (269). Subsequently, vRNPs assemble themselves as a structure of eight RNPs with one of each viral segment; this complex interacts with the M1 monolayer located at the budding area in the envelope. This initiates the budding process with proper packaging of eight RNA segments (270). NA is needed to release fully formed virions from the cell surface because these virions are covered with sialic acids and, hence, aggregate, and/or adhere to the cell surface. NA removes HA-sialic acid links, releasing viral particles (251,271). Release of the newly formed viral particles yields between 5 log and 6 log viral particles per milliliter of nasal mucus.

Antiviral Agents for Influenza

Given the incidence of influenza causing significant global morbidity and mortality, approaches to manage seasonal outbreaks mandate both therapeutic and public health interventions. However, influenza virus is continually evolving; it may either be altered through small gradual changes (antigenic drift) or abrupt major changes (antigenic shift) due to genetic reassortments, leading to resistant strains. Influenza resistance increases the risk of recurrence of worldwide pandemics, as in 1918 and 2009 (272). Since the introduction of amantadine in 1962, it was realized that therapeutic interventions could control and prevent such devastation (Fig. 11.6).

Anti-M2 Products Adamantines: Amantadine and Rimantadine

The first description of antiinfluenza activity of the M2 channel blocker amantadine was reported in 1962. Both amantadine and its derivative, rimantadine, inhibit in vitro replication of influenza A viruses at concentrations achievable in vivo. The target site is the M2 ion channel protein transmembrane domain (262). The ion channel activity is blocked by amantadine. As a result, M1 protein does not dissociate from the vRNPs blocking viral uncoating within the host cell, which is necessary for replication (246,250). Amantadine has a second, late effect on some subtypes of influenza A virus whose HA undergoes fusion-conformational changes at relatively high pH. Amantadine can block the H⁺ transport from the Golgi. This can lead to premature conformational changes of the HA from lack of control of the pH in this intracellular compartment.

As with all RNA viruses, influenza virus mutates relatively frequently. Mutations provide selective advantages, including resistance to antiviral drugs as well as to vaccines, which need to be updated annually due to this fact (259). As demonstrated by analysis of resistant mutants, altered M2 proteins containing a change in the transmembrane domain display resistant phenotype to amantadine and rimantadine. Sequencing of RNA segment 7 from resistant isolates has demonstrated that nucleotide changes in the transmembrane domain, such as single nucleotide changes at residues 26, 27, 30, 31, and 34, can result in a resistant phenotype (246). Among clinical cases, the most frequent resistant mutation observed is in codon 31. All amino acid residues involved in resistance are located in the drug binding site. Influenza BM2 protein is intrinsically resistant to adamantine therapy (252,261).

The use of amantadine or rimantadine results in the frequent emergence of resistant strains (246,273,274). Current strains of influenza A H3N2 and H1N1 are already resistant to these agents (274). Resistance can be observed by day 2 of use, and in up to 45% of treated patients by day 7. These resistant isolates have been found to be as contagious and as pathogenic as the wild-type strain. By 2005 to 2006, more than 90% of seasonal influenza strains were resistant to the adamantines (275). Given the high frequency of resistance, there is rarely a role to use these agents to treat influenza nowadays. The CDC no longer recommends use of either of these drugs (244,249).

Neuraminidase Inhibitors: Zanamivir and Oseltamivir

The most commonly prescribed influenza treatment nowadays target NA since the decline in use of the M2 inhibitors due to high resistance. Influenza NA functions as a sialidase and has a role in the release of the newly assembled virions from infected cells (246,268). This enzyme cleaves the linkage between a terminal sialic acid and the adjacent D-galactose or D-galactosamine. It is responsible for (a) the removal of sialic acids and (b) the release of budding virions. It also enables the transport of the virus through the mucin layer of the respiratory tract to bind to the target epithelial cell. The NA is a homotetramer, and each monomer harbors one catalytic site. This catalytic site is highly conserved among the different A subtypes and the B viruses.

It has long been postulated that sialic acid analogs may function as inhibitors to sialidases and may have antiviral activity (246,268,276,277). This competitive inhibition of NA may result in the inhibition of virus replication and spreading. Since 1999, new NA inhibitors have been licensed for use in several parts of the world (259,268,278). These licensed compounds display very effective inhibition of the NA enzymatic activity. The antiviral NA inhibitor interacts with amino acid residues of the catalytic pocket, especially glutamine 119, aspartate 151, arginine 152, glutamine 227, alanine 246, glutamine 276, and arginine 292. NA inhibitors bind the active site of influenza A and B viruses and prevent the cleavage of sialic acid molecules on the host cell membrane, efficiently preventing an important step in viral propagation (279) (Figs. 11.7 and 11.8). Mutations observed in these residues can lead to resistance (277) via subtype-specific mutations in framework or catalytic segments of NA (279).

Viruses with altered susceptibility to NA inhibitors have been identified recently (249,259,280). Given subtle variations in NA structures, resistance can be unique to individual NA inhibitors (274). Viruses with mutations in position 119, 152, 274, and 292 were resistant. In some viruses, mutations were also associated with changes in the HA protein (278), which acts synergistically with NA (259) and is near the RBS. These mutants lead to a reduced affinity to sialic acids and ease virion release when the sialidase activity of the NA is impaired because of mutations. However, reduced binding of HA may occur only for a specific subset of receptors. Receptors on Madin-Darby canine kidney (MDCK) cells, which are most often used for influenza virus growth inhibition assays, have both α-2,3– and α-2,6–linked terminal sialic acid residues, whereas human cells have only α-2,6–linked sialic acids (250). Hence, this system may not detect a virus with decreased binding capacities to α-2,6–linked sialic acids. In influenza A (H3N2) viruses, oseltamivir-resistant variants with the R292K mutation are reported. Drug-resistant strains from patients treated with oseltamivir have a mutation in E119V. The predominant resistance mutation against oseltamivir and peramivir in the influenza A (H1N1) virus is H274Y (274,279). By 2008 to 2009, more than 95% of seasonal H1N1 in the United States was resistant to oseltamivir on this basis (281). This mutant strain is still sensitive to zanamivir (274). A recent study created resistant strains using reverse genetic techniques to study drug resistance of various mutations, confirming these clinical observations (279). Multidrug-resistant H1N1 viruses may be due to certain NA mutations such as E119G and E119V. This severe form of resistance may produce increased risk for immunocompromised patients. It is thought that the potential for viral transmission is low, however (279). The I222V mutation increases resistance to oseltamivir as well as peramivir and raises concern about the potential public health implications of this double mutant strain of influenza (279).

A recent review found that oral oseltamivir reduces mortality in high-risk populations. Both oseltamivir and inhaled zanamivir also appear to reduce hospitalizations and decrease the duration of influenza symptoms (282). Oseltamivir-resistant strains predispose patients to develop pneumonia more often, particularly in immunocompromised individuals (272), increasing their morbidity and mortality.

The therapeutic regimen of neuraminidase inhibitors (NAIs) is shown in Table 11.18. Zanamivir is administrated by oral or nasal aerosol. It is poorly absorbed (4% to 17%). It has a half-life of 3.4 hours and it has a bioavailability of 10% to 20% (283). After administration of a single 10-mg dose by using an inhaling device to administer the micronized formulation, the peak is observed at 2.5 hours with a concentration of 34 µg/L, and 78% of the active product is deposited in the oropharynx. As measured in different studies, the pharmacokinetic parameters after a single inhaled dose of 10 mg were area under curve (AUC), 247 µg/L • hour; t_max, 0.75 hours; t_1/2, 3.56 hours; C_max, 39 µg/L; and Cl_ren, 6.45 L/hour. Zanamivir is not metabolized, and 90% is eliminated unmetabolized in the urine.

Oseltamivir is given by mouth. Oseltamivir phosphate is an ethyl ester prodrug that is hydrolyzed by hepatic esterases to its active form, oseltamivir, specifically, oseltamivir carboxylate (see Fig. 11.8). Absolute bioavailability is approximately 80%, with an elimination half-life of oseltamivir carboxylate of 6.7 to 8.2 hours. The prodrug is absorbed at 75%, with this form bound to serum proteins at 43%. Studies in healthy volunteers who received oral oseltamivir, 150 mg, and an intravenous infusion of oseltamivir, 150 mg, showed that the absolute bioavailability of oseltamivir (prodrug) was 79%, with a t_maxof 5 hours and a maximum concentration of 456 µg/L. The metabolite (active drug) had an AUC of 6,834 µg/L • hour, t_max of 2.88 hours, V_d of 25.6 liters, t_1/2 of 1.79 hours, C_max of 2,091 µg/L, Cl of 6.67 L/hour, and a Cl_ren of 18.8 L/hour, with approximately 93% excreted in the urine. Pharmacokinetics of the active metabolite appeared to be linear up to dose of 500 mg oseltamivir orally twice daily (283).

Peramivir is the only intravenous (IV) option used for the treatment of severe influenza. Given its IV formulation, peramivir is quite useful for patients with oseltamivir-resistant strains but who are unable to use inhaled zanamivir due to preexisting asthma. In October 2009, during the pandemic, the FDA gave an emergency use authorization (EUA) to allow administration of peramivir based on safety data from phase 1, 2, and 3 clinical trials. The EUA for peramivir use in the United States expired in June 2010. It should be noted that peramivir is in clinical use in Japan and South Korea (259,272).

Both zanamivir and oseltamivir are well tolerated, with only minor side effects (284). However, because of the rare report of exacerbation of asthma in some young asthmatic patients, oseltamivir is not indicated in children with chronic asthma.

Other Antiinfluenza Agents and Approaches

The high rate of mutation of the influenza virus, combined with its ability to intermix its multiple RNA gene segments, allow for rapid development of new strains. This allows the virus to resist vaccination as well as standard therapies. Hence, there is an ongoing need to seek and develop new antiinfluenza agents using innovative strategies. There are several new inhibitors that attack traditional and also novel targets of influenza virus replication (see Fig. 11.6).

Laninamivir is a long-acting NA inhibitor that is effective against oseltamivir-resistant viruses (259). Laninamivir is available as a single-dose inhalation therapy. Another investigational drug against NA is favipiravir (6-fluoro-3-hydroxy-2-pyrazinecarboxamide). As this drug is metabolized, it gains nucleoside activity and inhibits influenza-specific RNA polymerase but does not inhibit human polymerases. This agent is effective against seasonal influenza viruses as well as oseltamivir-sensitive or oseltamivir-resistant highly pathogenic H5N1 strains and adamantine-resistant strains of influenza. Favipiravir is undergoing clinical trials in Japan (285).

New therapies based on pathogenic mechanisms of influenza viral life cycle and infectivity, similar to the work on ARVs in the HIV section, includes efforts to produce entry blocker, or EB. This is derived from fibroblast growth factor 4, which specifically binds to influenza viral HA protein. This drug exhibited broad antiviral activity against influenza viruses and also prevented reinfection (259).

DAS181 (Fludase) is a fungal sialidase chimeric protein that is fused to a cell surface–binding domain. This molecule functions like viral NA and destroys surface receptors required for host cell sialic acid binding. In studies, DAS181 (Fludase) inhibited both human and avian influenza (259).

Ribavirin is a synthetic nucleoside analog used most often in the treatment of HCV and has in vitro antiviral activity against influenza A and B viruses (246,286,287). It inhibits cellular inosine-5′-monophosphate dehydrogenase activity, thereby depleting intracellular pools of guanosine 5′-triphosphate. The phosphorylated ribavirin has a direct inhibitory effect on viral RNA-dependent RNA polymerase activity. However, studies focused on the role of ribavirin, either intravenous or inhaled, showed no definitive clinical benefit. Although ribavirin does have in vitro activity against influenza, the risks of side effects such as hemolytic anemia and teratogenicity belie its potential benefit and make this agent unsuitable for the treatment of influenza (244).

Testing Methodology for Influenza Therapies and Resistance

The rapid emergence of amantadine/rimantadine resistance in the late 1990s (275) and the recent observations of rapidly developing resistance to NAI during the 2007 to 2008 flu season (281) focused on the necessity to monitor susceptibility to influenza therapies. During 2007 to 2008, oseltamivir-resistant seasonal influenza A (H1N1) emerged, wherein the viruses carrying the H275Y mutation in NA did not respond to this therapy. By the 2008 to 2009 season, more than 95% of seasonal influenza A (H1N1) carried this resistance gene (279,280). Potential for the development and spread of resistant variants to each new therapy reinforces this stance.

Testing of Anti-M2 Products

The initial assay described was a PRA, which depended on incubation interval and showed great variability in results (288). Hence, alternative methods that can be easily standardized have been developed (289,290). These techniques are rapid, as compared with PRAs, and use an automated colorimetric assay with antibodies that detect virus growth in the inoculated cells (290). Briefly, virus stocks of the tested strains are prepared in cells to obtain a virus titer of approximately 10⁴ per milliliter. Each virus stock is subsequently prepared in serial dilutions ranging from 10⁻¹ to 10⁻⁵; these dilutions are tested against six antiviral concentrations ranging from 40 to 0.0026 µg/mL in fivefold dilutions. The assay is performed in a 96-well microtiter plate and uses a chess-board titration technique that allows simultaneous titration of the virus both in the absence and the presence of increasing antiviral doses (amantadine or rimantadine). Vero or MDCK cells can be used; the results obtained with both cells are similar.

The titrated virus (i.e., 10⁴ per milliliter) is used for cell inoculation in 96-well microtiter plates. The microtiter plate is prepared by inoculating 15,000 cells per well. The plates are incubated for 2 days at 35°C with 5% CO₂. Following incubation, culture medium is removed and 25 µL of the antiviral dilution, 25 µL of the calibrated virus suspension, and 200 µL of EMEM plus trypsine (final concentration, 2 µg/mL) are added to each well. Each test is performed in duplicate. After inoculation, the plates are centrifuged at 225 × g for 30 minutes at room temperature. The plates are subsequently incubated at 35°C in a 5% CO₂ atmosphere for 20 hours for MDCK cells and 44 hours for Vero cells. Subsequently, the medium is removed from each well and the cells are fixed with 200 µL of 0.1% glutaraldehyde prepared in PBS. The fixation is performed for 15 minutes at 20°C. The wells are then washed with 500 µL of PBS. Virus protein detection is performed with a rabbit polyclonal antiserum to the A or B strain and is used to detect a wide range of influenza proteins. Briefly, 50 µL of rabbit antisera prepared in PBS plus 0.5% bovine serum albumin (BSA) is added in each well and incubated for 90 minutes at 35°C. The wells are then washed with 500 µL of PBS. Next, 50 µL of protein A horseradish peroxidase conjugate is added and the mixture is incubated for 60 minutes at 35°C. The wells are subsequently washed with PBS and 100 µL of the substrate solution is added to reveal the reaction. The substrate is a 2,2-azino-di-(3-ethylbenzthiazoline) sulfonic acid prepared in ABTS buffer. The incubation is performed at 20°C for 60 minutes. The optical densities are read at 405 nm by using a multichannel spectrophotometer; the data provided are analyzed with a microcomputer. For each antiviral concentration, a curve is drawn with the optical density values obtained with the different virus concentrations. This curve is compared with the control without antiviral agents. The antiviral concentration providing a 50% reduction in the production of antigenic material as determined at the optimal virus suspension is the IC₅₀ value. This value is calculated from the data provided with the tested antiviral concentrations.

For strains sensitive to rimantadine, the IC₅₀ range is from 0.02 to 1 µg/mL. The resistant strains have an IC₅₀ ranging from 4 to 28 µg/mL.

Testing of Neuraminidase Inhibitors

Several assays are available and have been evaluated for testing NA inhibitors (Table 11.19). Cell-based assays are not convenient because of the presence of alternative virus receptors to the α-2,6–linked sugar that is exclusively observed in human cells (291). Enzyme inhibition assays that are not subject to receptor specificity have been developed to determine enzyme inhibition with a substrate that mimics sialic acids. The most widely used substrate is the fluorogenic reagent 2′MUNANA (methyl umbelliferone N-acetyl aquaminic acid), initially described by Potier et al. (291).

The drawback of using fluorescent enzymatic assays is that evaluation of strains with poor enzymatic activity can be difficult. Therefore, an alternative chemiluminescent assay has been developed to overcome this problem (292), with a 1,2-dioxetane derivative of sialic acid. Both assays must be performed with the active antivirals zanamivir and oseltamivir carboxylate. Both procedures are summarized in Figure 11.9.

Fluorometric Neuraminidase Assay Method.The fluorometric assay was first described by Potier et al. (291). The input virus used for the assay must be titrated by making serial twofold dilutions and then graphically determining the virus and enzyme concentrations, which fall into the linear part of the curve for the inhibition assay. A signal-to-noise ratio of 2 or more is considered optimal for use in the inhibition assay. Equal volumes of drug and the appropriate virus dilution are mixed and incubated for 15 minutes at room temperature on 96-well microtiter plates. The final drug concentration uses ranges from 0.0033 to 30,000 nM in serial 1:10 dilutions. The reaction is initiated by the addition of 100 µM MUNANA substrate solution prepared in 2-(N- morpholino) ethanesulfonic acid (MES) buffer CaCl₂(32.5 mM MES, pH 6.5, 4 mM CaC1₂). After 1 hour of incubation (37°C, with shaking), the reaction is stopped by the addition of 150 µL of a freshly prepared 50 mM glycin solution whose pH has been equilibrated to 10.4 with NaOH. This buffer is preferred to an alternative 0.14 M NaOH buffer prepared in 83% ethanol, which can result in precipitates in the reacting well.

Influenza virus NA activity is assayed in the fluorometric test by using a modification of the technique described by Potier et al. (291). The MUN, 2′-(4-methylumbelliferyl)-α-D-N-acetylneuraminic acid, sodium salt (MW, 489.4), was used as substrate, and 32.5 mM MES [2-(N-morpholino) ethanesulfonic acid, sodium salt (MW, 217.2), pH 5.8, 4 mM CaC1₂] was chosen for the buffer. The 100-µM working solution of MUN was prepared in MES.

Detailed Neuraminidase Activity Assay.Reaction mixtures containing 25 µL of calibrated virus obtained by twofold dilutions, 25 µL of MES CaC1₂, and 50 µL of 100 µM MUN working solution are incubated in flat-bottom plates for 1 hour at 37°C. The reaction is terminated by the addition of 150 µL of 50 mM glycine buffer, pH 10.4.

Fluorescence was quantified in a fluorometer with an excitation wavelength of 355 nm and an emission wavelength of 460 nm. Relative fluorescent units of sample were measured and corrected using the mean value of blank controls obtained when carrying out the reaction in the absence of virus.

The quantitation of NA activity was deduced from comparison of sample values with a standard curve established by using 4-methylumbelliferone (4-Mu) sodium salt. The equation establishing the correlation between free 4-Mu concentration and the amount of fluorescence is written as follows:

A (nmol 4 − Mu/hour/mL) = 22.7 10⁻³ × Δ Fluorescence × Dilution⁻¹

(The constant value derived from the slope of the line.)

Detailed Neuraminidase Activity Inhibition Assay.A standard dose of virus (10 nmol/hour/mL) in 25 µL of MES buffer is preincubated with 25 µL of inhibitor dilution for 15 minutes at 37°C with shaking. After addition of 50 µL of 100 µM MUN working solution, the reaction mixture was incubated for 60 minutes at 37°C, with shaking. Finally, the reaction was terminated by adding 150 µL of 50 mM glycine buffer, pH 10.4. The IC₅₀ values of inhibitors were further calculated as the inhibitor concentration required for reducing NA activity by 50%.

When used to determine NA activity inhibition assay, the values obtained with sensitive and resistant isolates were easy to interpret (see Fig. 11.9). Values of IC₅₀ are between 0.05 and 2 nM, with the resistant isolates displaying IC₅₀ values beyond 50 nM.

Chemiluminescent Assay.The chemiluminescent assay was developed as an alternative to the fluorometric assay. It is described by Buxton et al. (292). It requires the growth of cells and virus in phenol red–free media, since residual phenol red can interfere with the assay. The virus is initially titrated in twofold dilutions in 32.5 MES (pH 6.0)-4 mM CaCl₂, and the signal-to-noise ratio of 40 is used for the dilution. Final drug concentrations ranged from 0.028 to 550 nM in serial 1:3 dilutions. The appropriate viral NA dilution (40 µL) was preincubated with 10 µL of drug for 30 minutes at room temperature on white Optiplates (Packard, Meridan, CT). The reaction was started by the addition of 5 µL of a 1:9 dilution of NA-Star prepared in 32.5 mM MES (pH 6.0)-4 mM CaC1₂. The final concentration of substrate used in the assay was 100 µM. The reaction mixture was incubated at 37°C for 15 minutes, with shaking. Chemiluminescent light emission was triggered by the addition of 55 µL of Light Emmission Accelerator II (Applied Biosystems, Foster City, CA) to each well. The half-life of the mixture is 5 minutes, which means that rapid automation is required to optimize sensitivity. All assay results were immediately read with an Applied Biosystems’ (Foster City, CA) NORTHSTAR Luminometer.

IC₅₀ values were calculated by using the Robosage Microsoft Excel software add-in for curve fitting and calculation of IC₅₀ values. The equation used for the calculation of IC₅₀ values was y = V_max × (1 − [x/K + x]). This equation describes a simple hyperbolic inhibition curve with a zero baseline. In this equation, x is the inhibitor concentration, y is the response being inhibited (i.e., the velocity of an enzymatic reaction), and V_max is the limiting response as x approaches zero. As x increases without bound, y tends toward its lower limit, zero. K is the IC₅₀ for the inhibition curve; that is, y is 50% V_max when x is equal to K. Graphic presentations of the raw data are used to identify possible sources of variability and to characterize any patterns in the data (see Fig. 11.9).

The results obtained with both chemiluminescent and fluorometric assays are consistent regarding the values for resistance thresholds in influenza A and B viruses. The IC₅₀ values of the resistant isolates are beyond 20 nM for influenza A stains and beyond 40 nM for influenza B. Some differences are observed between IC₅₀ values obtained with oseltamivir carboxylate and zanamivir.

Procedure for Testing of the Susceptibility of Influenza A Viruses to Neuraminidase Inhibition Fluorometric Neuraminidase Assay Method

Influenza viruses are susceptible to NA inhibition. The fluorometric assay, described here, begins with the collection of nasal swab specimens. This procedure may be adapted to begin with nasal washes or throat swab specimens. Virus isolation must be performed on MDCK cells.

A. Isolation and Culture of Influenza Virus from Clinical Specimens

Clinical isolates are obtained by the culture of virus from clinical specimens on MDCK cells. Alternative culture systems for virus detection that have been proposed are less efficient than the MDCK cell system.

1.  MDCK Cell Culture.MDCK cells can be purchased (ATCC, CCL34). Cells can be routinely passaged twice weekly in serum-free Ultra-MDCK medium (Cambrex Bioscience, Walkersville, MD) supplemented with 2 mM L-glutamine, penicillin (225 U/mL), and streptomycin (225 µg/mL). Alternatively, the passage medium can be EMEM with the same supplementation and with 2 µg/mL trypsin.

2.  Virus Isolates.Swab samples are removed from the transport tube and expressed in 2.5 mL of EMEM, which has been supplemented with the antibiotic mix (penicillin plus streptomycin). The transport medium is then added. This suspension is inoculated onto MDCK subconfluent monolayers in 24-well microtiter plates. To promote growth of the influenza viruses, the microtiter plates are subsequently centrifuged at 300× g for 30 minutes (293). The cells are then incubated at 33°C in 5% CO₂ atmosphere. At day 4, the monolayers are checked for cytopathic effect. The supernatant is collected and may be tested for HA activity or blindly passaged. The HA activity of the supernatant must reach a minimum titer of 16 to detect NA activity. At least two passages are necessary to obtain this titer on MDCK cell culture supernatants.

3.  Virus Controls.In each assay, a reference-sensitive and reference-resistant isolate should be included. The resistant isolate should have a specific resistant mutation in the NA gene (i.e., position R152). The susceptible isolates used were the ongoing vaccine prototype strains. When testing an isolate, the resistant and susceptible controls are from the same subtype. These reference strains are tested in the same conditions as the isolates.

B. Neuraminidase Activity Assay

Influenza virus NA activity is assayed in the fluorometric test by using a modification of the technique previously described by Potier et al. (291). The assay measures 4-methylumbelliferone released from the fluorogenic substrate 2′-MUN by the influenza virus NA.

1.  Reagents.The buffer is 32.5 mL of 100 mM MES sodium salt, pH 5.8, and 4 mL of 100 mM CaCl₂ made up to a 100-mL final volume with sterile water. Store at 4°C.

The substrate mix is 100-µM working solution of MUN prepared in 32.5 mM MES, pH 5.8, 4-mM CaCl₂ buffer. Store 500-µL aliquots of 5 mM MUN at −20°C and dilute 1:50 in 32.5 mM MES, pH 5.8, 4 mM CaCl₂ buffer immediately prior to assay. This substrate mix must be stored in the dark (wrapped in aluminum foil).

The stop solution is 50 mM glycine, pH 10.4. Store at 4°C.

2.  Titration of Neuraminidase Activity.The assay is performed in 96-well flat-bottom plates to allow optical reading from the bottom. Prior to performing the titration, creation of a standard curve is necessary to control the linearity of the assay within the concentration range used. This is also performed in a 96-well flat-bottom plate.

Standard Curve

Free 4-methylumbelliferone can be used to generate a standard curve and to verify the linearity of the assay. The reagents are 4-methylumbelliferone, sodium salt, and a stop solution of 50 mM glycine, pH 10.4. Store at 4°C.

For the reagent preparation of 5 mM of 4-methylumbelliferone stock solution, dissolve 9.91 mg in 10 mL of water. Store at 4°C in the dark for up to 2 months. For the preparation of 10 µM of 4-methylumbelliferone standard solution, dissolve 20 µL of stock solution into 10 mL of 50 mM glycine buffer, pH 10.4. Store at 4°C in the dark for up to 2 months. With the standard stock solution, prepare the dilution from 0.2 to 5 µM. Fluorescence is quantified with a fluorometer with an excitation wavelength of 355 nm and an emission wavelength of 460 nm.

Testing of Isolates

The isolate testing is done in 96-well flat-bottom plates. The reagents are added as follows: 25 µL of serial twofold dilutions (from 1/2 to 1/2,048) of culture supernatant of the controls and the tested isolates (HA titer above 16) prepared in 32.5 mM MES, pH 5.8; 4 mM CaCl₂ buffer are added in columns 2 to 12. Then 25 µL of 32.5 mM MES, pH 5.8, and 4 mM CaCl₂ buffer are added to each well, except for the well of column 1 where the volume is 50 µL (reaction control). Fifty microliters of substrate are then added in every well to start the reaction. (A freshly made 100-µM working solution of MUN must be prepared in 32.5 mM MES, pH 5.8, and 4 mM CaCl₂ buffer.)

The contents of the plates are mixed gently on a mechanical vibrator. The plates are incubated for 1 hour at 37°C. The reaction is then stopped by adding 150 µL of 50 mM glycine buffer, pH 10.4, in each well. Fluorescence can be measured with a fluorometer with an excitation wavelength of 355 nm and an emission wavelength of 460 nm. If the measure is delayed, the plate must be stored in the dark.

Relative fluorescent units of a sample are measured and corrected using the mean value of blank controls obtained when performing the reaction in the absence of the virus (column 1). The quantification of NA activity is deduced from comparison of sample values with a standard curve established by using 4-Mu sodium salt. The equation establishing the correlation between free 4-Mu concentration and the amount of fluorescence is as follows:

A (nmol 4 − Mu/hour/mL) = 22.7 10⁻³ × Δ Fluorescence × Dilution⁻¹

(The constant value is derived from the slope of the line.) The A value obtained by this calculation is used to determine the input of the virus in the NA activity inhibition assay. This input must be 10 nmoL/hour/mL.

3.  Neuraminidase Activity Inhibition Assay.The assay is done in 96-well flat-bottom plates to allow optical reading from the bottom. Reagents are added as described:

Twenty-five microliters of 10-fold dilutions of NA inhibitors prepared in 32.5 mM MES, pH 5.8, 4 mM CaCl₂ buffer (from 30,000 to 0.003 nM) are added in the wells from columns 1 to 8. Column 9 is used to determine the blank value and contains 50 µL of 32.5 mM MES, pH 5.8, and 4 mM CaCl₂ buffer. Twenty-five microliters of a standard dose of virus or control (10 nmoL/hour/mL, as determined in the titration of NA activity) prepared in 32.5 mM MES, pH 5.8, 4 mM CaCl₂ buffer is added to each well except well 9. Wells 10 to 12 are used as a control of the standard dose.

The contents of the plates are gently mixed on a mechanical vibrator. The plates are incubated for 15 minutes at 37°C. Then 50 µL of substrate is added in each well to start the reaction. (A freshly made 100-µM working solution of MUN is prepared in 32.5 mM MES, pH 5.8, and 4 mM CaCl₂ buffer.) The contents of the plates are gently mixed on a mechanical vibrator and the plates are incubated for 1 hour at 37°C. The reaction is then stopped by adding 150 µL of 50 mM glycine buffer, pH 10.4, in all wells. Fluorescence is immediately quantified by using a fluorometer with an excitation wavelength of 355 nm and an emission wavelength of 460 nm.

If this measurement is delayed, the plates must be stored in the dark. The IC₅₀ values of antiviral agents are calculated according to the concentration required for reducing NA activity by 50%.

Phenotypic Assays

In general, genotypic assays are largely replaced by phenotypic assays for the detection of resistance to adamantine derivatives, amantadine and rimantadine. Complete cross-resistance between these antivirals is associated with amino acid substitution at position 26, 27, 30, 31, or 34 in the transmembrane region of the M2 protein of influenza A viruses (289). Conventional (Sanger) sequencing and pyrosequencing of the NA gene has also been used (294). The most frequently reported mutation with oseltamivir H1N1 resistance is H274Y (295). Mutations for E19V, R292K, and N294S were associated with oseltamivir resistance in N2NA (294).

It is crucial to develop new agents/approaches to increase the global antiinfluenza armamentarium (259,285). There are five ongoing antiinfluenza approaches being used.

 1.  First is the development of new drugs that are directed against well-known viral proteins, such as NA, HA, the NS1 (e.g., laninamivir), and new agents directed against M2 ion channels.

 2.  Identification of effective compounds (see Fig. 11.6) that reduce virus growth has been identified that utilize small interfering RNA (siRNA) sequences to target influenza genes and limit viral protein synthesis. Investigational siRNA products targeting various influenza viral genes have been tested over the last few years but concerns regarding compound delivery and agent stability have yet to be resolved (259).

 3.  The third approach includes innovative treatments that target host factors, which are essential for virus replication, such as the host sialic acid receptor using DAS181 (Fludase) as noted earlier (259,285).

 4.  Fourth, multidrug combination therapies that target several viral functions at once are being studied (285). Triple therapy with oseltamivir, amantadine, and ribavirin has demonstrated increased effectiveness against influenza A in high-risk populations; randomized clinical trials are underway to confirm these preliminary data. Other combinations of antiinfluenza agents will likely be studied as well given these preliminary observations.

 5.  Finally, the fifth possibility for new antiinfluenza regimens is to target naturally occurring innate peptides and other host immunologic factors, such as defensins, antibodies, and interferons, to maximize their antiinfluenza effects (259,285). Nitazoxanide, originally used as an antiparasitic agent, has been found to upregulate interferon response to influenza but also to exhibit specific inhibitory effect on the maturation of the HA protein (285). Defensins, also called retrocyclins, convey glycoprotein-binding inhibition, preventing fusion of other viruses, such as HIV, with host cells, revealing a novel antiviral mechanism. Another defensin may inhibit HA-mediated influenza viral fusion as well. Recently, retrocyclin synthetic analogs have been able to block influenza infection, suggesting that modification of defensins may produce new therapeutic agents to treat influenza (259). Antibody infusions derived from convalescent plasma have been used. Specific monoclonal antibodies (MABs) have also been developed. These MABs block HA from binding to its receptor, limiting the conformational change of the cell membrane, hence preventing infection. Several nonrandomized studies using antibodies have demonstrated decreased mortality in patients with complicated influenza (285). Clinical trials are under development to make use of these varied immunologic approaches.

Influenza remains a significant pathogen, which is the focus of global public health efforts. The rapid mutability of this virus and development of resistance genes has made worldwide control a challenge. Innovative therapies relying on both viral and host mechanisms are in use in some countries and other novel approaches are in development. Careful monitoring of viral resistance patterns and drug sensitivities will continue to be an important effort in the coming years to prevent and manage future pandemics.

  1.  National Committee on Clinical Laboratory Standards. Antiviral susceptibility testing: herpes simplex virus by plaque reduction assay; approved standard. M11-A4. Wayne, PA: National Committee for Clinical Laboratory Standards, 2004.

  2.  Harmenberg J, Wahren B, Oberg B. Influence of cells and virus multiplicity on the inhibition of herpesviruses with acycloguanosine. Intervirology 1980;14(5-6):239–244.

  3.  Baldanti F, Underwood, MR, Talarico CL, et al. The Cys 607→Tyr change in the UL97 phosphotransferase confers ganciclovir resistance to two human cytomegalovirus strains recovered from two immunocompromised patients. Antimicrob Agents Chemother 1998;42:444–446.

  4.  Richman DD, Guatelli JC, Grimes J, et al. Detection of mutations associated with zidovudine resistance in human immunodeficiency virus utilizing the polymerase chain reaction. J Infect Dis 1991;164:1073–1081.

  5.  St. Clair MH, Martin JL, Tudor-Williams G, et al. Resistance to ddI and sensitivity to AZT induced by a mutation in HIV-1 reverse transcriptase. Science 1991;253:1557–1559.

  6.  Biron KK, Fyfe JA, Stanat SC, et al. A human cytomegalovirus mutant resistant to the nucleoside analog 9-{[2-hydroxy-1-(hydroxymethyl) ethoxy]} guanine (BW B579U) induces reduced levels of BW B759U triphosphate. Proc Natl Acad Sci USA 1986;83:8769–8773.

  7.  Gadler H. Nucleic acid hybridization for measurement of effects of antiviral compounds on human cytomegalovirus DNA replication. Antimicrob Agents Chemother 1983;24:370–374.

  8.  Belshe RB, Burk B, Newman F, et al. Resistance of influenza A virus to amantadine and rimantadine: results of one decade of surveillance. J Infect Dis 1989;159:430–435.

  9.  Bright RA, Medina MJ, Xu X, et al. Incidence of adamantine resistance among influenza A (H3N2) viruses isolated worldwide from 1994-2005: a cause for concern. Lancet 2005;366:1175–1181.

 10.  Deyde VM, Xu X, Bright RA, et al. Surveillance of resistance to adamantanes among influenza A(H3N2) and A(H1N1) viruses isolated worldwide. J Infect Dis 2007;196:249–257.

 11.  Durant J, Clevenbergh P, Halfon P, et al. Drug-resistance genotyping in HIV-1 therapy: the VIRADAPT randomised controlled trial. Lancet 1999; 353(9171):2195–2199.

 12.  Hirsch MS, Brun-Vezinet F, Clotet B, et al. Antiretroviral drug resistance testing in adults infected with human immunodeficiency virus type 1: 2003 Recommendations of an International AIDS Society–USA Panel. Clin Infect Dis 2003;37:113–128

 13.  The Euroguidelines Group for HIV resistance. Clinical and laboratory guidelines for the use of HIV-1 drug resistance testing as part of treatment management: recommendations for the European setting. AIDS 2001;15:309–320.

 14.  Demeter L, Haubrich RJ. Phenotypic and genotypic resistance assays: methodology, reliability, and interpretations. J Acquir Immune Defic Syndr 2001;26(Suppl 1):S3–S9.

 15.  Shafer RW. Genotypic testing for human immunodeficiency virus type 1 drug resistance. Clin Microbiol Rev 2002;2:247–277.

 16.  Vandamme AM, Houyez F, Banhegyi D, et al. Laboratory guidelines for the practical use of HIV drug resistance tests in patient follow-up. Antivir Ther 2001;6:21–39.

 17.  Shafer RW, Warford A, Winters MA, et al. Reproducibility of human immunodeficiency virus type 1 (HIV-1) protease and reverse transcriptase sequencing of plasma samples from heavily treated HIV-1 infected individuals. J Virol Methods 2000;86:143–153.

 18.  Japour AJ, Fiscus SA, Arduino JM, et al. Standardized microtiter assay for determination of syncytium-inducing phenotypes of clinical human immunodeficiency virus type 1 isolates. J Clin Microbiol 1994;32:2291–2294.

 19.  Brun-Vézinet F, Ingrand D, Deforges L, et al. HIV-1–1 sensitivity to zidovudine: a consensus culture technique validated by genotypic analysis of the reverse transcriptase. J Virol Methods 1992;37:177–188.

 20.  Kellam P, Larder BA. Recombinant virus assay: a rapid, phenotypic assay for assessment of drug susceptibility of human immunodeficiency virus type 1 isolates. Antimicrob Agents Chemother 1994;38:23–30.

 21.  Petropoulos CJ, Parkin NT, Limoli KL, et al. A novel phenotypic drug susceptibility assay for human immunodeficiency virus type 1. Antimicrob Agents Chemother 2000;44:920–928.

 22.  Whitcomb JM, Huang W, Fransen S, et al. Analysis of baseline Enfuvirtide (T20) susceptibility and co-receptor tropism in two-phase III study populations. Proceedings of 10th conference on retrovirus and opportunistic infections, Boston, MA, Feb 10–14, 2003; Abstract 557.

 23.  Qari SH, Respess R, Weinstock H, et al. Comparative analysis of two commercial phenotypic assays for drug susceptibility testing of human immunodeficiency virus type 1. J Clin Microbiol 2002;40:31–35.

 24.  Japour AJ, Mayers DL, Johnson VA, et al. Standardized peripheral blood mononuclear cell culture assay for determination of drug susceptibilities of clinical human immunodeficiency virus type 1 isolates. Antimicrob Agents Chemother 1993;37:1095–1101

 25.  Hertogs K, DeBethune MP, Miller V, et al. A rapid method for simultaneous detection of phenotypic resistance to inhibitors of protease and reverse transcriptase in recombinant human immunodeficiency virus type 1 isolates in patients treated with antiretroviral drugs. Antimicrob Agents Chemother1998;42:269–276.

 26.  Kempf DJ, Isaacson JD, King MS, et al. Analysis of the virological response with respect to baseline viral phenotype and genotype in protease inhibitor-experienced HIV-1-infected patients receiving lopinavir/ritonavir therapy. Antivir Ther 2002;7:165–174.

 27.  Lanier ER, Hellmann N, Scott J, et al. Determination of a clinically relevant phenotypic resistance cut-off for abacavir using the PhenoSense assay. In: Proceedings of the 8th Conference on Retrovirus and Opportunistic Infections; February 4–8, 2001; Chicago, IL. Abstract 254.

 28.  Margot NA, Isaacson E, McGowan I, et al. Genotypic and phenotypic analyses of HIV-1 in antiretroviral-experienced patients treated with tenofovir DF. AIDS 2002;16:1227–1235.

 29.  Katzenstein DA, Bosch RJ, Wang N, et al. Baseline phenotypic susceptibility and virological failure over 144 weeks among NRTI experienced subjects in ACTG 364. In: Proceedings of the 9th Conference on Retrovirus and Opportunistic Infections; February 24–28, 2002; Seattle, WA. Abstract 591.

 30.  Shulman NS, Hughes MD, Winters MA, et al. Subtle decreases in stavudine phenotypic susceptibility predict poor virologic response to stavudine monotherapy in zidovudine-experienced patients. J Acquir Immune Defic Syndr 2002;31:121–127.

 31.  Whitcomb JM, Huang W, Limoli K, et al. Hypersusceptibility to nonnucleoside reverse transcriptase inhibitors in HIV-1: clinical, phenotypic and genotypic correlates. AIDS 2002;16:F41–F47.

 32.  Katzenstein DA, Bosch RJ, Hellmann N, et al. Phenotypic susceptibility and virological outcome in nucleoside-experienced patients receiving three or four antiretroviral drugs. AIDS 2003;17:821–830.

 33.  Haubrich R, Keiser P, Kemper C, et al. CCTG 575: a randomised, prospective study of phenotype versus standard of care for patients failing antiretroviral therapy. Antivir Ther 2001;6(Suppl 1):63.

 34.  Sanger F, Nicklen S, Coulson AR. DNA sequencing with chain terminating inhibitors. Proc Natl Acad Sci USA 1977;74:5463–5467.

 35.  Meynard JL, Vray M, Morand-Joubert L, et al. Phenotypic or genotypic resistance testing for choosing antiretroviral therapy after treatment failure: a randomized trial. AIDS 2002;16:727–736.

 36.  Brun-Vézinet F, Descamps D, Ruffault A, et al. Clinically relevant interpretation of genotype for resistance to abacavir. AIDS 2003;17:1795–1802.

 37.  Calvez V, Costagliola D, Descamps D, et al. Impact of stavudine phenotype and thymidine analogue mutations on viral response to stavudine plus lamivudine in ALTIS 2 ANRS trial. Antivir Ther 2002;7:211–218.

 38.  Shulman NS, Machekano RA, Shafer RW, et al. Genotypic correlates of a virologic response to stavudine after zidovudine monotherapy. J Acquir Immune Defic Syndr 2001;27:377–380.

 39.  Descamps D, Masquelier B, Mamet JP, et al. A genotypic sensitivity score for amprenavir based on genotype at baseline and virological response. Antivir Ther 2001;6:S59.

 40.  Masquelier B, Breilh D, Neau D, et al. Human immunodeficiency virus type 1 genotypic and pharmacokinetic determinants of the virological response to lopinavir-ritonavir-containing therapy in protease inhibitor-experienced patients. Antimicrob Agents Chemother2002;46:2926–2932.

 41.  Calvez V, Cohen-Codar I, Marcelin AG, et al. Identification of individual mutations in HIV protease associated with virological response to lopinavir/ritonavir therapy. Antivir Ther 2001;6 (Suppl 1):64.

 42.  Van Laethem K, De Luca A, Antinori A, et al. A genotypic drug resistance interpretation algorithm that significantly predicts therapy response in HIV-1-infected patients. Antivir Ther 2002;7:123–129.

 43.  Tural C, Ruiz L, Holtzer C, et al. Clinical utility of HIV-1 genotyping and expert advice: the Havana trial. AIDS 2002;216:209–218.

 44.  Ravela J, Betts BJ, Brun-Vezinet F, et al. HIV-1 protease and reverse transcriptase mutation patterns responsible for discordances between genotypic drug resistance interpretation algorithms. J Acquir Immune Defic Syndr 2003;33:8–14.

 45.  De Luca A, Cingolani A, Di Giambenedetto S, et al. Variable prediction of antiretroviral treatment outcome by different systems for interpreting genotypic human immunodeficiency virus type 1 drug resistance. J Infect Dis 2003;187:1934–1943.

 46.  Schmidt B, Walter H, Schwingel E, et al. Comparison of different interpretation systems for genotypic HIV-1 drug resistance data. Antivir Ther 2001;6(Suppl 1):102.

 47.  Larder BA, Kemp SD, Hertogs K. Quantitative prediction of HIV-1 phenotypic drug resistance from genotypes: the virtual phenotype (Virtual Phenotype). Antivir Ther 2000;5(Suppl 3):49–50.

 48.  Perez-Elias MJ, Garcia I, Munoz V. A randomized prospective study of phenotype (P) versus virtual phenotype (virtualP) testing for patients failing antiretroviral therapy. In: Proceedings of the 42nd Interscience Conference on Antimicrobial Agents; September 27–30, 2002; San Diego, CA. Abstract H-1079.

 49.  Parkin N, Chappey C, Maroldo L, et al. Phenotypic and genotypic HIV-1 drug resistance assays provide complementary information. J Acquir Immune Defic Syndr 2002;31:128–136.

 50.  Brun-Vézinet F, Race E, Descamps D, et al. Difference between genotype and phenotype in the Narval trial, ANRS 088. Antivir Ther 2000;5(Suppl 3):78.

 51.  Bacheler L, Jeffrey S, Hanna G, et al. Genotypic correlates of phenotypic resistance to efavirenz in virus isolates from patients failing nonnucleoside reverse transcriptase inhibitor therapy. J Virol 2001;75:4999–5008.

 52.  Falloon J, Ait-Khaled M, Thomas DA, et al. HIV-1 genotype and phenotype correlate with virological response to abacavir, amprenavir and efavirenz in treatment-experienced patients. AIDS 2002;16:387–396.

 53.  Hsu A, Isaacson J, Brun S, et al. Pharmacokinetic-pharmacodynamic analysis of lopinavir-ritonavir in combination with efavirenz and two nucleoside reverse transcriptase inhibitors in extensively pretreated human immunodeficiency virus-infected patients. Antimicrob Agents Chemother 2003;47:350–359.

 54.  Shulman N, Zolopa A, Havlir D, et al. Virtual inhibitory quotient predicts response to ritonavir boosting of indinavir-based therapy in human immunodeficiency virus-infected patients with ongoing viremia. Antimicrob Agents Chemother 2002;46:3907–3916.

 55.  Marcelin AG, Lamotte C, Delaugerre C, et al. Genotypic inhibitory quotient as predictor of virological response to ritonavir-amprenavir in human immunodeficiency virus type 1 protease inhibitor-experienced patients. Antimicrob Agents Chemother 2003;47:594–600.

 56.  Perelson AS, Neumann AU, Markovitz M, et al. HIV-1 dynamics in vivo: virion clearance rate, infected cell life-span, and viral generation time. Science 1996;271:1582–1586.

 57.  Weinstein MC, Goldie SJ, Losina E, et al. Use of genotypic resistance testing to guide HIV therapy: clinical impact and cost effectiveness. Ann Intern Med 2001;134:440–450.

 58.  Schuurman R, Demeter L, Reichelderfer P, et al. Worldwide evaluation of DNA sequencing approaches for identification of drug resistance mutations in the human immunodeficiency virus type 1 reverse transcriptase. Clin Microbiol 1999;37:2291–2296.

 59.  Larder BA, Kohli A, Kellam P, et al. Quantitative detection of HIV-1 drug resistance mutations by automated DNA sequencing. Nature 1993;365:671–673.

 60.  Gunthard HF, Wong JK, Ignacio CC, et al. Comparative performance of high density oligonucleotide sequencing and dideoxynucleotide sequencing of HIV type 1 pol from clinical samples. AIDS Res Hum Retroviruses 1998;14:869–876.

 61.  D’Aquila RT. Limits of resistance testing. Antivir Ther 2000;5:71–76.

 62.  Charpentier C, Dwyer DE, Lecossier D, et al. Co-evolution and competition of viral populations with distinct resistance genotypes in patients failing treatment with protease inhibitors. Antivir Ther 2002;7:S42.

 63.  Moser MJ, Sharma PL, Nurpeisov V, et al. Using GENE-CODE technology to detect early emerging populations of drug-resistant human immunodeficiency virus Type 1. In: Proceedings of the XII International HIV Drug Resistance Workshop: Basic Principles and Clinical Implications; June 10–14, 2003; Los Cabos, Mexico. Abstract 89.

 64.  Simmonds P, Zhang LQ, McOmish F, et al. Discontinuous sequence change of human immunodeficiency virus (HIV) type 1 env sequences in plasma viral and lymphocyte-associated proviral populations in vivo: implications for models of HIV pathogenesis. J Virol1991;65:6266–6276.

 65.  Smith MS, Koerber KL, Pagano JS. Zidovudine-resistant human immunodeficiency virus type 1 genomes detected in plasma distinct from viral genomes in peripheral blood mononuclear cells. J Infect Dis 1993;167:445–448.

 66.  Wei X, Ghosh SK, Taylor ME, et al. Viral dynamics in human immunodeficiency virus type 1 infection. Nature 1995;373:117–122.

 67.  Koch N, Yahi N, Ariasi F, et al. Comparison of human immunodeficiency virus type 1 (HIV-1) protease mutations in HIV-1 genomes detected in plasma and in peripheral blood mononuclear cells from patients receiving combination drug therapy. J Clin Microbiol1999;37:1595–1597.

 68.  Tashima KT, Flanigan TP, Kurpewski J, et al. Discordant human immunodeficiency virus type 1 drug resistance, including K103N, observed in cerebrospinal fluid and plasma. Clin Infect Dis 2002;35:82–83.

 69.  Ghosn J, Viard JP, De Almeida M, et al. HIV-1 resistance pattern may be frequently different than in blood in the male genital compartment of patients with therapeutic failure. Antivir Ther 2002;7:S46.

 70.  Solas C, Lafeuillade A, Halfon P, et al. Discrepancies between protease inhibitor concentrations and viral load in reservoirs and sanctuary sites in human immunodeficiency virus-infected patients. Antimicrob Agents Chemother 2003;47:238–243.

 71.  Marlowe N, Swanson P, Drews B, et al. Performance of Celera Diagnostics ViroSeq HIV-1 genotyping system on genetically diverse HIV-1 strains. In: Proceedings of the 10th Conference on Retroviruses and Opportunistic Infections; February 10–14, 2003; Boston, MA. Abstract 582.

 72.  Jagodzinski LL, Cooley JD, Weber M, et al. Performance characteristics of human immunodeficiency virus type 1 (HIV-1) genotyping systems in sequence-based analysis of subtypes other than HIV-1 subtype B. J Clin Microbiol 2003;41:998–1003.

 73.  Beddows S, Galpin S, Kazmi SH, et al. Performance of two commercially available sequence-based HIV-1 genotyping systems for the detection of drug resistance against HIV type 1 group M subtypes. J Med Virol 2003;70:337–342.

 74.  Kantor R, Katzenstein DA, Efron B, et al. Impact of HIV-1 subtype and antiviral therapy on protease and reverse transcriptase genotype: results of a global collaboration. PLoS Med 2005;2:325–337.

 75.  Beasley RP, Lin C-C, Hwang LY, et al. Hepatocellular carcinoma and hepatitis B virus: a prospective study of 22,707 men in Taiwan. Lancet 1981;2:1129–1133.

 76.  Rehermann B, Fowler P, Sidney J, et al. The cytotoxic T lymphocyte response to multiple hepatitis B virus polymerase epitopes during and after acute viral hepatitis. J Exp Med 1995;181:1047–1058.

 77.  Rehermann B, Lau D, Hoofnagle JH, et al. Cytotoxic T lymphocyte responsiveness after resolution of chronic hepatitis B virus infection. J Clin Invest 1996;97:1655–1665.

 78.  Guidotti LG, Chisari FV. To kill or to cure: options in host defense against viral infection. Curr Opin Immunol 1996;8:478–483.

 79.  Ganem D, Schneider R. Hepadnaviridae: the viruses and their replication. In: Knipe DM, Howley PM, eds. Fields virology. Philadelphia: Lippincott Williams & Wilkins, 2001;2923–2970.

 80.  Kann M, Gerlich W. Hepadnaviridae: structure and molecular virology. In: Zuckerman A, Thomas H, eds. Viral hepatitis. London: Churchill Livingstone, 1998;77–105.

 81.  Kao JH. Clinical relevance of hepatitis B viral genotypes: a case of déjà vu? J Gastroenterol Hepatol 2002;17:113–115.

 82.  Kao JH. Hepatitis B viral genotypes: clinical relevance and molecular characteristics. J Gastroenterol Hepatol 2002;17:643–650.

 83.  Kao JH, Liu CJ, Chen DS. Hepatitis B viral genotypes and lamivudine resistance. J Hepatol 2002;36:303–304.

 84.  Sugauchi F, Orito E, Ichida T, et al. Hepatitis B virus of genotype B with or without recombination with genotype C over the precore region plus the core gene. J Virol 2002;76:5985–5992.

 85.  Owiredu WK, Kramvis A, Kew MC. Hepatitis B virus DNA in serum of healthy black African adults positive for hepatitis B surface alone: possible association with recombination between genotypes A and D. J Med Virol 2001;64:441–454.

 86.  Will H, Reiser W, Weimer T, et al. Replication strategy of human hepatitis B virus. J Virol 1987;61:904–911.

 87.  Okamoto H, Imai M, Kametani M, et al. Genomic heterogeneity of hepatitis B virus in a 54-year-old woman who contracted the infection through maternal fetal transmission. Jpn J Exp Med 1987;57:231–236.

 88.  Gunther S, Fischer L, Pult I, et al. Naturally occurring variants of hepatitis B virus. Adv Virus Res 1999;52:25–137.

 89.  Zhang YY, Summers J. Enrichment of a precore-minus mutant of duck hepatitis B virus in experimental mixed infections. J Virol 1999;73:3616–3622.

 90.  Zhang YY, Summers J. Low dynamic state of viral competition in a chronic avian hepadnavirus infection. J Virol 2000;74:5257–5265.

 91.  Doo E, Liang TJ. Molecular anatomy and pathophysiologic implications of drug resistance in hepatitis B virus infection. Gastroenterology 2001;120:1000–1008.

 92.  Seeger C, Mason WS. Hepatitis B virus biology. Microbiol Mol Biol Rev 2000;64:51–68.

 93.  Lok AS, Akarca U, Greene S. Mutations in the pre-core region of hepatitis B virus serve to enhance the stability of the secondary structure of the pre-genome encapsidation signal. Proc Natl Acad Sci USA 1994;91:4077–4081.

 94.  Liang TJ, Hasegawa K, Rimon N, et al. A hepatitis B virus mutant associated with an epidemic of fulminant hepatitis. N Engl J Med 1991;324:1705–1709.

 95.  Omata M, Ehata T, Yokosuka O, et al. Mutations in the precore region of hepatitis B virus DNA in patients with fulminant and severe hepatitis. N Engl J Med 1991;324:1699–1704.

 96.  Okamoto H, Yotsumoto S, Akahane Y, et al. Hepatitis B viruses with precore region defects prevail in persistently infected hosts along with seroconversion to the antibody against e antigen. J Virol 1990;64:1298–1303.

 97.  Hunt CM, McGill JM, Allen MI, et al. Clinical relevance of hepatitis B viral mutations. Hepatology 2000;31:1037–1044.

 98.  Akarca U, Lok AS. Naturally occurring hepatitis B virus core gene mutations. Hepatology 1995;31:1037–1044.

 99.  Sirma H, Gianninni C, Poussin K, et al. Hepatitis B virus X mutants, present in hepatocellular carcinoma tissue abrogate both the antiproliferative and transactivation effects of HBx. Oncogene 1999;18:4848–4859.

100.  Carman WF, Zanetti AR, Karayiannis P, et al. Vaccine-induced escape mutant of hepatitis B virus. Lancet 1990;336:325–329.

101.  Carman WF, Trautwein C, Van Deursen FJ, et al. Hepatitis B virus envelope variation after transplantation with and without hepatitis B immune globulin prophylaxis. Hepatology 1996;24:489–493.

102.  Tavis J. The replication strategy of the hepadnaviruses. Viral Hepatitis Rev 1996;2:205–218.

103.  Nassal M, Schaller H. Hepatitis B virus replication. J Viral Hepatitis 1996;3:217–226.

104.  Hu J, Toft D, Seeger C. Hepadnavirus assembly and reverse transcription require a multi-component chaperone complex which is incorporated into nucleocapsids. EMBO 1997;16:59–68.

105.  Von Weizacker F, Kock J, Weiland S, et al. Cis-preferential recruitment of duck hepatitis B virus core protein to the RNA/polymerase preassembly complex. Hepatology 2002;35:209–216.

106.  Lok AS, Heathcote EJ, Hoofnagle JH. Management of hepatitis B: summary of a workshop. Gastroenterology 2001;120:1828–1853.

107.  Guidotti LG, Rochford R, Chung J, et al. Viral clearance without destruction of infected cells during acute HBV infection. Science 1999;284:825–829.

108.  Thimme R, Wieland S, Steiger C, et al. CD8+ T cells mediate viral clerance and disease pathogenesis during acute hepatitis B virus infection. J Virol 2003;77:68–76.

109.  Guo JT, Zhou H, Liu C, et al. Apoptosis and regeneration of hepatocytes during recovery from transient hepadnavirus infections. J Virol 2000;74:1495–505.

110.  Kajino K, Jilbert AR, Saputelli J, et al. Woodchuck hepatitis virus infections: very rapid recovery after a prolonged viremia and infection of virtually every hepatocyte. J Virol 1994;68:5792–5803.

111.  Boni C, Bertoletti A, Penna A, et al. Lamivudine treatment can restore T cell responsiveness in chronic hepatitis B. J Clin Invest 1998;102:968–975.

112.  Valsamakis A. Molecular testing in the diagnosis and management of chronic hepatitis B. Clin Microbiol Rev 2007;20:426–439.

113.  Chevaliez S, Rodriguez C, Pawlotsky JM. New virologic tools for management of chronic hepatitis B and C. Gastroenterology 2012;142:1303–1313.

114.  Shaw T, Bartholomeusz A, Locarnini S. HBV drug resistance: mechanisms, detection and interpretation. J Hepatol 2006;44:593–606.

115.  Chao DC, Hu KQ. Update on rescue therapies in patients with lamivudine-resistant chronic hepatitis B. Drug Design Develop Ther 2013;7:777–788.

116.  Doong SL, Tsai CH, Schinazi RF, et al. Inhibition of the replication of hepatitis B virus in vitro by 2,′ 3′-dideoxy-3′-thiacytidine and related analogues. Proc Natl Acad Sci USA 1991;88:8495–8499.

117.  Liaw YF, Leung NW, Chang TT, et al. Effects of extended lamivudine therapy in Asian patients with chronic hepatitis B. Asia Hepatitis Lamivudine Study Group. Gastroenterology 2000;119:172–180.

118.  Leung NW, Lai CL, Chang TT, et al. Extended lamivudine treatment in patients with chronic hepatitis B enhances hepatitis e antigen seroconversion rates: results after 3 years of therapy. Hepatology 2001;33:1527–1532.

119.  Lewin SR, Ribeiro RM, Walters T, et al. Analysis of hepatitis B viral load decline under potent therapy: complex decay profiles observed. Hepatology 2001;34:1012–1020.

120.  Ono SK, Kato N, Shiratori Y, et al. The polymerase L528M mutation cooperates with nucleotide binding-site mutations, increasing hepatitis B virus replication and drug resistance. J Clin Invest 2001;107:449–455.

121.  Das K, Xiong X, Yang H, et al. Molecular modeling and biochemical characterization reveal the mechanism of hepatitis B virus polymerase resistance to lamivudine (3TC) and emtricitabine. J Virol 2001;75:4771–4779.

122.  Allen MI, Gauthier J, DesLauriers M, et al. Two sensitive PCR-based methods for detection of hepatitis B virus variants associated with reduced susceptibility to lamivudine. J Clin Microbiol 1999;37:3338–3347.

123.  Lau DTY, Khokhar F, Doo E, et al. Long-term therapy of chronic hepatitis B with lamivudine. Hepatology 2000;32:828–834.

124.  Naesens L, Snoeck R, Andrei G, et al. HPMPC (cidofovir), PMEA (adefovir) and related acyclic nucleoside phosphonate analogues: a review of their pharmacology and clinical potential in the treatment of viral infections. Antiviral Chem Chemother 1997;8:1–23.

125.  Heijtink RA, De Wilde GA, Kruining J, Berk L, et al. Inhibitory effect of 9-(2-phosphonylmethoxyethyl) adenine (PMEA) on human and duck hepatitis B virus infection. Antiviral Res 1993;21:141–153.

126.  Del Gobbo V, Foli A, Balzarini J, et al. Immunomodulatory activity of 9-(2-phosphonylmethoxyethyl) adenine (PMEA), a potent anti-HIV nucleoside analogue, on in vivo murine models. Antiviral Res 1991;16:65–75.

127.  Calio R, Villani N, Balestra E, et al. Enhancement of natural killer activity and interferon induction by different acyclic nucleoside phosphonates. Antiviral Res 1994;23:77–89.

128.  Westland CE, Yang H, Delaney IV WE, et al. Week 48 resistance surveillance in two phase 3 clinical studies of adefovir dipivoxil for chronic hepatitis B. Hepatology 2003;38:96–103.

129.  Angus P, Vaughan R, Xiong S, et al. Resistance to adefovir dipivoxil therapy associated with the selection of a novel mutation in the HBV polymerase. Gastroenterology 2003;125:292–297.

130.  Westland CE, Villeneuve JP, Terrault N, et al. Resistance surveillance of liver transplantation patients with lamivudine-resistant hepatitis B virus after 96 weeks of adefovir dipivoxil treatment. Hepatology 2003;38:160A.

131.  Balfour HHJ. Antiviral drugs. N Engl J Med 1999;340:1255–1268.

132.  Honkoop P, De Man RA. Entecavir: a potent new antiviral drug for hepatitis B. Expert Opin Investig Drugs 2003;12:683–688.

133.  Yamanaka G, Wilson T, Innaimo S, et al. Metabolic studies on BMS-200475, a new antiviral compound active against hepatitis B virus. Antimicrob Agents Chemother 1999;43:190–193.

134.  Levine S, Hernandez D, Yamanaka G, et al. Efficacies of entecavir against lamivudine-resistant hepatitis B virus replication and recombinant polymerases in vitro. Antimicrob Agents Chemother 2002;46:2525–2532.

135.  Marion PL, Salazar FH, Winters MA, et al. Potent efficacy of entecavir (BMS-200475) in a duck model of hepatitis B virus replication. Antimicrob Agents Chemother 2002;46:82–88.

136.  Colonno RJ, Genovesi EV, Medina I. Long term entecavir treatment results in sustained antiviral efficacy and prolonged life span in the woodchuck model of chronic hepatitis infection. J Infect Dis 2001;184:1236–1245.

137.  Wolters LMM, Hansen BE, Niesters HGM. Viral dynamics during and after entecavir therapy in patients with chronic hepatitis B. J Hepatol 2002;37:137–144.

138.  Tassopoulos NC, Hadziyannis SJ, Cianciara J, et al. Entecavir is effective in treating patients with chronic hepatitis B who failed lamivudine therapy. Hepatology 2001;34:340A.

139.  Hayden FG. Antimicrobial agents (continued) antiviral agents (Nonretroviral). In: Hardman JG, Limbird LE, eds, Gilman AG, consulting ed. Goodman & Gilman’s: the pharmacological basis of therapeutics, 10th ed. New York: McGraw-Hill, 2001:1313–1347.

140.  James JS. Tenofovir approved: broad indication. AIDS Treat News 2001;(373):2–3.

141.  Van Bommel F, Wunsche T, Schurmann D, et al. Tenofovir treatment in patients with lamivudine-resistant hepatitis B mutants strongly affects viral replication. Hepatology 2002;36:507–508.

142.  Le Guerhier F, Pichoud C, Guerret S, et al. Characterization of the antiviral effect of 2′3′-dideoxy-2′-3′ didehydro-beta-L-5-fluorocytidine in the duck hepatitis B virus infection model. Antimicrob Agents Chemother 2000;44:111–122.

143.  Balakrishna PSS, Liu H, Zhu YL. Inhibition of hepatitis B virus by a novel L-nucleoside, 2′ fluoro-5-methyl-beta-L-arabinofuranosyl uracil. Antimicrob Agents Chemother 1998;40:380–386.

144.  Aguesse-Germon S, Liu H, Chevalier M, et al. Inhibitory effect of 2′ fluoro-5-methyl-beta-L-arabinofuranosyl uracil on duck hepatitis B virus replication. Antimicrob Agents Chemother 1998;42:369–376.

145.  Pierra C, Imbach JL, De Clercq E, et al. Synthesis and antiviral evaluation of some beta-L-2′-3′ dideoxy-5-chloropyrimidine nucleosides and pronucleotides. Antiviral Res 2000;45:169–183.

146.  Zhou XJ, Lim SG, Lai CL, et al. Phase 1 dose escalation pharmacokinetics of L-deoxithymidine in patients with chronic hepatitis B virus infection. Hepatology 2001;34:629A.

147.  King RW, Ladner SK, Miller TJ, et al. Inhibition of human hepatitis B virus replication by AT-61, a phenyl-propenamide derivative, alone and in combination with (−)beta-L-2,′ 3′-dideoxy-3′thiacytidine. Antimicrob Agents Chemother 1998;42:3179–3186.

148.  Delaney WE IV, Edwards R, Colledge D, et al. The phenylpropenamide derivatives AT-61 and AT-130 inhibit replication of both wild-type and lamivudine resistant strains of hepatitis B virus in vitro. Antimicrob Agents Chemother 2002;46:3057–3060.

149.  Feld J, Colledge D, Sozzi T, et al. The phenylpropenamide derivative AT-130 inhibits HBV replication at viral encapsidation and packaging. Hepatology 2002;36:300A.

150.  Deres K, Schroder CH, Paessens A, et al. Inhibition of hepatitis B virus replication by drug-induced depletion of nucleocapsids. Science 2003;299:893–896.

151.  Kamiya N, Kubota A, Iwase Y, et al. Antiviral activities of MCC-478, a novel and specific inhibitor of hepatitis B virus. Antimicrob Agents Chemother 2002;46:2872–2877.

152.  Ono-Nita SK, Kato N, Shiratori Y, et al. Novel nucleoside analogue MCC-478 (LY582563) is effective against wild-type or lamivudine-resistant hepatitis B virus. Antimicrob Agents Chemother 2002;46:2302–2605.

153.  The European Association for the Study of Liver Jury. EASL international consensus conference on hepatitis B, 13–14 September, 2002, consensus statement. J Hepatol 2003;38:533–540.

154.  Blum HE. Gene therapy of viral hepatitis. In: Tsuji T, Higashi T, Zeniya M, et al, eds. Molecular biology and immunology in hepatology. Freiburg, Germany: Elsevier Science, 2002;97–109.

155.  Branch A. A hitchhiker’s guide to antisense and nonantisense biochemical pathways. Hepatology 1996;24:1517–1529.

156.  Hannon GJ. RNA interference. Nature 2002;41:244–251.

157.  Shlomai A, Shaul Y. Inhibition of hepatitis B virus expression and replication by RNA interference. Hepatology 2003;37:764–770.

158.  Lauer GM, Walker BD, MD. Hepatitis C virus infection. N Engl J Med 2001;345:41–52.

159.  Colombo M. Natural history and pathogenesis of hepatitis C virus related hepatocellular carcinoma. J Hepatol 1999;31:25–30.

160.  Penin F. Structural biology of hepatitis C virus. Clin Liver Dis 2003;7:1–21.

161.  Glue P, Rouzier-Panis R, Raffanel C, et al. A dose-ranging study of pegylated interferon alfa-2b and ribavirin in chronic hepatitis C. The Hepatitis C Intervention Therapy Group. Hepatology 2000;32:647–653.

162.  Fried MW, Shiffman ML, Reddy KR, et al. Peginterferon alfa-2a plus ribavirin for chronic hepatitis C virus infection. N Engl J Med 2002;347:975–982.

163.  Jensen DM, Marcellin B, Freilich B, et al. Re-treatment of patients with chronic hepatitis C who do not respond to peginterferon-alpha2b: a randomized trial. Ann Intern Med 2009;150:528–540.

164.  Jacobson IM, Gonzalez SA, Ahmed F, et al. A randomized trial of pegylated interferon alpha-2b plus ribavirin in the retreatment of chronic hepatitis C. Am J Gastroenterol 2005;100:2453–2462.

165.  Heathcote EJ, Shiffman ML, Cooksley WG, et al. Peginterferon alfa-2a in patients with chronic hepatitis C and cirrhosis. N Engl J Med 2000;343:1673–1680.

166.  Bressanelli S, Tomei L, Roussel A, et al. Crystal structure of the RNA-dependent RNA polymerase of hepatitis C virus. Proc Natl Acad Sci USA 1999;96:13034–13039.

167.  Tellinghuisen TL, Foss KL, Treadaway JC, et al. Identification of residues required for RNA replication in domain II and III of the hepatitis C virus NS5A protein. J Virol 2008;82:1073–1083.

168.  Kim JL, Morgenstern KA, Lin C, et al. Crystal structure of the hepatitis C virus NS3 protease domain complexed with a synthetic NS4A cofactor peptide. Cell 1996;87:343–355.

169.  Fox AN, Jacobsen IM. Recent successes and noteworthy future prospects in the treatment of chronic hepatitis C. Clin Infect Dis 2012;55(Suppl 1):S16–S24.

170.  Choo QL, Kuo G, Weiner AJ, et al. Isolation of a cDNA clone derived from a blood-borne non-A, non-B viral hepatitis genome. Science 1989;244:359–362.

171.  Robertson B, Myers G, Howard C, et al. Classification, nomenclature, and database development for hepatitis C virus (HCV) and related viruses: proposals for standardization. International Committee on Virus Taxonomy. Arch Virol 1998;143:2493–2503.

172.  Bantef H, Schulze-Ostoff K. Apoptosis in hepatitis C virus infection. Cell Death Differ 2003;10(Suppl 1):S48–S58.

173.  Pawlotsky JM, Germanidis G, Frainais PO, et al. Evolution of the hepatitis C virus second envelope protein hypervariable region in chronically infected patients receiving alpha interferon therapy. J Virol 1999;73:6490–6499.

174.  Simmonds P, Alberti A, Alter HJ, et al. A proposed system for the nomenclature of hepatitis C viral genotypes. Hepatology 1994;19:1321–1324.

175.  Wasley A, Alter MJ. Epidemiology of hepatitis C: geographic differences and temporal trends. Semin Liver Dis 2000;20:1–16.

176.  Martell M, Esteban JI, Quer J, et al. Hepatitis C virus (HCV) circulates as a population of different but closely related genomes: quasispecies nature of HCV genome distribution. J Virol 1992;66:3225–3229.

177.  Giannini C, Bréchot C. Hepatitis C virus biology. Cell Death Differ 2003;10:27–38.

178.  Pawlotsky JM. Hepatitis C virus resistance to antiviral therapy. Hepatology 2000;32:889–896.

179.  Bartenschlager R, Lohmarm V. Novel cell culture systems for the hepatitis C virus. Antiviral Res 2001;52:1–17.

180.  Griffin SD, Beales LP, Clarke DS, et al. The p7 protein of hepatitis C virus forms an ion channel that is blocked by the antiviral drug, amantadine. FEBS Lett 2003;535:34–38.

181.  Kaito M, Watanabe S, Tsukiyama-Kohara K, et al. Hepatitis C virus particle detected by immunoelectron microscopic study. J Gen Virol 1994;75:1755–1760.

182.  Thomssen R, Bonk S, Thiele A. Density heterogeneities of hepatitis C virus in human sera due to the binding of beta-lipoproteins and immunoglobulins. Med Microbiol Immunol 1993;182:329–334.

183.  Loriot MA, Bronowicki JP, Lagorce D, et al. Permissiveness of human biliary epithelial cells to infection by hepatitis C virus. Hepatology 1999;29:1587–1595.

184.  Moldvay J, Deny P, Pot S, et al. Detection of hepatitis C virus RNA in peripheral blood mononuclear cells of infected patients by in situ hybridization. Blood 1994;83:269–273.

185.  Pileri P, Uematsu Y, Campagnoli S, et al. Binding of hepatitis C virus to CD81. Science 1998;282:938–941.

186.  Meola A, Sbardellati A, Bruni Ercole B, et al. Binding of hepatitis C virus E2 glycoprotein to CD81 does not correlate with species permissiveness to infection. J Virol 2000;74:5933–5938.

187.  Flint M, Quinn ER, Levy S. In search of hepatitis C virus receptor(s). Clin Liver Dis 2001;5:873–893.

188.  Cocquerel L, Wychowski C, Minner F, et al. Charged residues in the transmembrane domains of hepatitis C virus glycoproteins play a major role in the processing, subcellular localization, and assembly of these envelope proteins. J Virol 2000;74:3623–3633.

189.  Cocquerel L, Op de Beeck A, Lambot M, et al. Topological changes in the transmembrane domains of hepatitis C virus envelope glycoproteins. EMBO J 2002;21:2893–2902.

190.  Beales LP, Rowlands DJ, Holzenburg A. The internal ribosome entry site (IRES) of hepatitis C virus visualized by electron microscopy. RNA 2001;7:661–670.

191.  Reed KE, Rice CM. Overview of hepatitis C virus genome structure, polyprotein processing, and protein properties. Curr Top Microbiol Immunol 2000;242:55–84.

192.  Kunkel M, Lorinczi M, Rijnbrand R, et al. Self-assembly of nucleocapsid-like particles from recombinant hepatitis C virus core protein. J Virol 2001;75(5):2119–2129.

193.  Fried MW, Shiffman ML, Reddy KR, et al. Peg interferon alfa-2a plus ribavirin for chronic hepatitis C virus infection. N Engl J Med 2002;347:975–982.

194.  Hay AJ, Wolstenholme AJ, Skehel JJ, et al. The molecular basis of the specific anti-influenza action of amantadine. EMBO J 1985;4:3021–3024.

195.  Lau JY, Tam RC, Liang TJ, et al. Mechanism of action of ribavirin in the combination treatment of chronic HCV infection. Hepatology 2002;35:1002–1009.

196.  Dellamonica P, Calvez V, Marcelin AG, et al. The Antivirogram and the modes of action of antiviral agents, HIV, hepatitis, influenza and CMV. In: Lorian V, ed. Antibiotics in Laboratory Medicine. Philadelphia: Lippincott Williams & Wilkins, 2005;564–614

197.  Mercer DF, Schiller DE, Elliott JF, et al. Hepatitis C virus replication in mice with chimeric human livers. Nat Med 2001;7:927–933.

198.  Castet V, Fournier C, Soulier A, et al. Alpha interferon inhibits hepatitis C virus replication in primary human hepatocytes infected in vitro. J Virol 2002;76:189–199.

199.  Rumin S, Berthillon P, Tanaka E, et al. Dynamic analysis of hepatitis C virus replication and quasispecies selection in long-term cultures of adult human hepatocytes infected in vitro. J Gen Virol 1999;80:3007–3018.

200.  Lohmann V, Komer F, Koch J, et al. Replication of subgenomic hepatitis C virus RNAs in a hepatoma cell line. Science 1999;285:110–113.

201.  Blight KJ, Kolykhalov AA, Rice CM. Efficient initiation of HCV RNA replication in cell culture. Science 2000 Dec 8;290:1972–1974.

202.  Piertschmann T, Lohmann V, Rutter G, et al. Characterization of cell lines carrying self-replicating hepatitis C virus RNAs. J Virol 2001;75:1252–1264.

203.  Pawlotsky JM. Mechanisms of antiviral treatment efficacy and failure in chronic hepatitis C. Antiviral Res 2003;59:1–11.

204.  Lamarre D, Anderson PC, Bailey M, et al. An NS3 protease inhibitor with antiviral effects in humans infected with hepatitis C virus. Nature 2003;426:186–189.

205.  Harrington PR, Zeng W, Neger LK. Clinical relevance of detectable but quantifiable hepatitic C virus RNA during boceprevir or telaprevir treatment. Hepatology 2012;55:1048–1057.

206.  Perlman BL, Traub N. Sustained virologic response to antiviral therapy for chronic hepatitis C virus infection: a cure and so much more. Clin Infect Dis 2011;52:889–900.

207.  U.S. Food and Drug Administration. Chronic hepatitis C virus infection: developing direct-acting antiviral agents for treatment. Draft guidance for industry. September, 2010.

208.  Ferenci P, Laferl H, Scherzer TM, et al. PEG-interferon alfa-2a and ribavirin for 24 weeks in hepatitis C type 1 and 4 patients with rapid virological response. Gastroenterology 2008;135:451–458.

209.  Ghany MG, Strader DB, Thomas DL, et al. Diagnosis, management, and treatment of hepatitis C: an update. Hepatology 2009;49:1335–1374.

210.  Bowen EF, Emery VC, Wilson P, et al. Cytomegalovirus polymerase chain reaction viraemia in patients receiving ganciclovir maintenance therapy for retinitis. AIDS 1998;12:605–611.

211.  Boivin G, Gilbert C, Gaudreau A, et al. Rate of emergence of cytomegalovirus (CMV) mutations in leukocytes of patients with acquired immunodeficiency syndrome who are receiving valganciclovir as induction and maintenance therapy for CMV retinitis. J Infect Dis2001;184:1598–1602.

212.  Weinberg A, Jabs DA, Chou S, et al. for the cytomegalovirus retinitis and viral resistance study group and the adult AIDS clinical trials group cytomegalovirus laboratories. Mutations conferring foscarnet resistance in a cohort of patients with acquired immunodeficiency syndrome and cytomegalovirus retinitis. J Infect Dis 2003;187:777–784.

213.  Limaye AP, Corey L, Koelle DM, et al. Emergence of ganciclovir-resistant cytomegalovirus disease among recipients of solid-organ transplants. Lancet 2000;356:645–649.

214.  Ducancelle A, Belloc S, Alain S, et al. Comparison of sequential cytomegalovirus isolates in a patient with lymphoma and failing antiviral therapy. J Clin Virol 2004;29:241–247.

215.  Lurain NS, Weinberg A, Crumpacker CS, et al. Sequencing of cytomegalovirus UL97 gene for genotypic antiviral resistance testing. Antimicrob Agents Chemother 2001;45:2775–2780.

216.  Landry ML, Stanat S, Biron K, et al. A standardized plaque reduction assay for determination of drug susceptibilities of cytomegalovirus clinical isolates. Antimicrob Agents Chemother 2000;44:688–692.

217.  Dankner WM, Scholl D, Stanat SC, et al. Rapid antiviral DNA-DNA hybridization assay for human cytomegalovirus. J Virol Methods 1990;28:293–298.

218.  McSharry JM, Lurain NS, Drusano GL, et al. Flow cytometric determination of ganciclovir susceptibilities of human cytomegalovirus clinical isolates. J Clin Microbiol 1998;36:958–964.

219.  Tatarowicz WA, Lurain NS, Thompson KD. In situ ELISA for the evaluation of antiviral compounds effective against human cytomegalovirus. J Virol Methods 1991;35:207–215.

220.  Pépin JM, Simon F, Dussault A, et al. Rapid determination of human cytomegalovirus susceptibility to ganciclovir directly from clinical specimens primocultures. J Clin Microbiol 1992;30:2917–2920.

221.  Gerna G, Sarasini A, Percivalle E, et al. Rapid screening for resistance to ganciclovir and foscarnet of primary isolates of human cytomegalovirus from culture-positive blood samples. J Clin Microbiol 1995;33:738–741.

222.  Eckle T, Prix L, Jahn G, et al. Drug-resistant human cytomegalovirus infection in children after allogeneic stem cell transplantation may have different clinical outcomes. Blood 2000;96:3286–3289.

223.  Chou E, Lurain NS, Thompson KD, et al. Viral DNA polymerase mutations associated with drug resistance in human cytomegalovirus. J Infect Dis 2003;188:32–39.

224.  Baldanti F, Michel D, Simoncini L, et al. Mutations in the UL97 ORF of ganciclovir-resistant clinical cytomegalovirus isolates differentially affect GCV phosphorylation as determined in a recombinant vaccinia virus system. Antiviral Res 2002;54:59–67.

225.  Sullivan V, Talarico CL, Stanat SC, et al. A protein kinase homologue controls phosphorylation of ganciclovir in human cytomegalovirus-infected cells. Nature 1992;358:162–164.

226.  Smith IL, Cherrington JM, Jiles RE, et al. High-level resistance of cytomegalovirus to ganciclovir is associated with alterations in both the UL97 and DNA polymerase genes. J Infect Dis 1997;176:69–77.

227.  Chou E, Marousek G, Guentzel S, et al. Evolution of mutations conferring multidrug resistance during prophylaxis and therapy for cytomegalovirus disease. J Infect Dis 1997;176:786–789.

228.  Chou E, Marousek G, Parenti DM, et al. Mutation in region III of the DNA polymerase gene conferring foscarnet resistance in cytomegalovirus isolates from three subjects receiving prolonged antiviral therapy. J Infect Dis 1998;178:526–530.

229.  Cihlar T, Fuller MD, Cherrington J. Characterization of drug resistance-associated mutations in the human cytomegalovirus DNA polymerase gene by using recombinant mutant viruses generated from overlapping DNA fragments. J Virol 1998;72:5927–5936.

230.  Erice A. Resistance of human cytomegalovirus to antiviral drugs. Clin Microbiol Rev 1999;12(Suppl 2):S286–S297.

231.  Sullivan V, Biron KK, Talarico CL, et al. A point mutation in the human cytomegalovirus DNA polymerase gene confers resistance to ganciclovir and phosphonylmethoxyalkyl derivatives. Antimicrob Agents Chemother 1993;37:19–25.

232.  Hirt B. Selective extraction of polyoma DNA from infected mouse cell cultures. J Mol Biol 1967;26:365–369.

233.  Alain S, Mazeron MC, Pepin JM, et al. Rapid detection of cytomegalovirus strains resistant to ganciclovir through mutations within the gene UL 97. Mol Cell Probes 1993;7:487–495.

234.  Chou S, Guentzel S, Michels KR, et al. Frequency of UL97 phosphotransferase mutations related to ganciclovir resistance in clinical cytomegalovirus isolates. J Infect Dis 1995;172:239–242.

235.  Chou S, Erice A, Jordan MC, et al. Analysis of the UL97 phosphotransferase coding sequence in clinical cytomegalovirus isolates and identification of mutations conferring ganciclovir resistance. J Infect Dis 1995;171:576–583.

236.  Hanson MH, Preheim LC, Chou S, et al. Novel mutation in the UL97 gene of a clinical cytomegalovirus strain conferring resistance to ganciclovir. Antimicrob Agents Chemother 1995;39:1204–1205.

237.  Harada K, Eizuru Y, Isashiki Y, et al. Genetic analysis of a clinical isolate of human cytomegalovirus exhibiting resistance against both ganciclovir and cidofovir. Arch Virol 1997;142:215–225.

238.  Chou S, Lurain NS, Weinberg A, et al. Interstrain variation in the human cytomegalovirus DNA polymerase sequence and its effect on genotypic diagnosis of antiviral drug resistance. Antimicrob Agents Chemother 1999;43:1500–1502.

239.  Fillet AM, Auray L, Alain S, et al. Natural polymorphism of cytomegalovirus DNA polymerase lies in two non-conserved regions spanning between domains delta-C and II and between domains III and I. Antimicrob Agents Chemother 2004;48:1865–1868.

240.  Kotton CN, Kumar D, Caliendo AM, et al. Updated international consensus guidelines on the management of cytomegalovirus in solid-organ transplantation. Transplantation 2013;96:333–360.

241.  Zambon MC. Epidemiology and pathogenesis of influenza. J Antimicrobial Chemother 1999;44:3–9.

242.  World Health Organization. Influenza virus infections in humans. http://www.who.int/influenza/GIP_InfluenzaVirusInfectionsHumans_Jul13.pdf?ua=1. Accessed December 15, 2013 .

243.  Scholtissek C. Source for influenza pandemics. Eur J Epidemiol 1994;10:455–458.

244.  Labella AM, Merel SE. Influenza. Med Clin North Am 2013;97:621–645.

245.  Reid AH, Fanning TG, Hultin JV, et al. Origin and evolution of the 1918 “Spanish” influenza gene. Proc Natl Acad Sci USA 1999;96:1651–1656.

246.  Stuart-Harris CH, Schild GC, Oxford JS, eds. Influenza. The virus and the disease. London: Edward Arnold, 1985:103–117.

247.  Aymard M, Valette M, Lina B, et al. Surveillance and impact of influenza in Europe. Vaccine 1999;17:S30–S41.

248.  Schrauwen EJ, de Graaf M, Herfst S, et al. Determinants of virulence of influenza A virus. Eur J Clin Microbiol Infect Dis 2013; Sep: 29.

249.  Fiore AE, Fry A, Shay D, et al. Antiviral agents for the treatment and chemoprophylaxis of influenza—recommendation for the Advisory Committee on Immunization Practices (ACIP). MMWR Recomm Rep 2011;60:1–24.

250.  Murphy BR, Webster RG. Orthomyxoviruses. In: Fields BN, Knipe DM, Howley PM, eds. Virology. 3rd ed. Lippincott-Raven: Philadelphia, 1996:1397–1446.

251.  Samji T. Influenza A: understanding the viral life cycle. Yale J of Biol and Med 2009;82:153–159.

252.  Farrukee R, Mosse J, Hurt AC. Review of the clinical effectiveness of neuraminidase inhibitors against influenza B viruses. Expert Rev Anti Infect Ther 2013;11:1135–1145.

253.  Towers S, Chowell G, Hameed R, et al. Climate change and influenza: the likelihood of early and severe influenza seasons following warmer than average winters. PLoS Curr 2013;5.

254.  Zambon MC. The pathogenesis of influenza in humans. Rev Med Virol 2001;11:227–241.

255.  Tong HH, Long JP, Shannon A, et al. Expression of cytokine and chemokine genes by human middle ear epithelial cells induced by influenza A virus and Streptococcus pneumoniae opacity variants. Infect Immun 2003;71:4289–4296.

256.  Joseph C, Togawa Y, Shindo N. Bacterial and viral infections associated with influenza. Influenza Other Respir Viruses 2013;7(Suppl 2):105–113.

257.  Harper SA, Bradley JS, Englund JA, et al. Seasonal influenza in adults and children—diagnosis, treatment, chemoprophylaxis and institutional outbreak management: clinical practice guidelines of the Infectious Diseases Society of America. Clin Infect Dis 2009;48:1003–1032.

258.  Centers for Disease Control and Prevention. Influenza antiviral medications: summary for clinicians. http://www.cdc.gov/flu/pdf/professionals/antivirals/antiviral-summary-clinicians.pdf. Accessed December 15, 2013.

259.  Barik S. New treatments for influenza. BMC Med 2012;10:104.

260.  Prevention and control of influenza with vaccines: recommendations of the Advisory Committee on Immunization Practices (ACIP)—United States, 2012–13 influenza season. http://www.cdc.gov/mmwr/preview/mmwrhtml/. Accessed December 15, 2013.

261.  Pielak RM, Chou JJ. Influenza M2 proton channels. Biochim Biophys Acta 2011;1808:522–529.

262.  Holsinger LJ, Nichari D, Pinto LH, et al. Influenza A virus M2 ion channel protein: a structure-function analysis. J Virol 1994;68:1551–1563.

263.  Whiley DC, Skehel JJ. The structure and function of the hemagglutinin membrane glycoprotein of influenza virus. Annu Rev Biochem 1987;56:365–394.

264.  Skehel JJ, Wiley DC. Receptor binding and membrane fusion in virus entry: the influenza hemagglutinin. Annu Rev Biochem 2000;69:531–569.

265.  Voeten JTM, Bestebroer TM, Nieuwkoop NJ, et al. Antigenic drift in the influenza A virus (H3N2) nucleoprotein and escape from recognition by cytotoxic T lymphocytes. J Virol 2000;74:6800–6807.

266.  Wiley DC, Wilson IA, Skehel JJ. Structural identification of the antibody-binding sites of Hong Kong influenza haemagglutinin and their involvement in antigenic variation. Nature 1981;289:373–378.

267.  Zhirnov OP, Ikizler MR, Wright PF. Cleavage of influenza A virus hemagglutinin in human respiratory epithelium is cell associated and sensitive to exogenous antiproteases. J Virol 2002;76:8682–8689.

268.  Coleman PM. Influenza virus neuraminidase: structure, antibodies, and inhibitors. Protein Sci 1994;3:1687–1696.

269.  Barman S, Ali A, Hui EK, et al. Transport of viral proteins to the apical membranes and interaction of matrix protein with glycoproteins in the assembly of influenza viruses. Virus Res 2001;77:61–69.

270.  Kawoka J. Enigmas of influenza. S4–2. Paper presented at: Fifth Annual Options for the control of Influenza Conference; October 7–11, 2003; Okinawa, Japan.

271.  Blick TJ, Sahasrabudhe A, Mcdonald M, et al. The interaction of neuraminidase and haemagglutinin mutations in the influenza virus in resistance to 4-guanadino-Neu5Ac2en. Virology 1998;246:95–103.

272.  Thorlund K, Awad T, Boivin G, et al. Systematic review of influenza resistance to the neuraminidase inhibitors. BMC Infectious Dis 2011;11:134–147.

273.  Dolin R, Reichmein RC, Madore HP, et al. A controlled trial of amantadine and rimantadine in the prophylaxis of influenza A infection. N Engl J Med 1982;307:580–584.

274.  McKimm-Breschkin JL. Influenza neuraminidase inhibitors: antiviral action and mechanisms of resistance. Influenza Other Respir Viruses 2013;7(Suppl 1):25–36.

275.  Bright RA, Shay DK, Shu B, et al. Adamantane resistance among influenza A viruses isolated early during the 2005-2006 influenza season in the United States. JAMA 2006;295:891–894.

276.  Palese P, Compans RW. Inhibition of influenza virus replication in tissue culture by 2-deoxy-2,3-dehydro-N-trifluoro-acetylneuraminic acid (FANA): mechanism of action. J Gen Virol 1976;33:159–163.

277.  McKimm-Breschin JL. Resistance of influenza viruses to neuraminidase inhibitors: a review. Antiviral Res 2000;47:1–17.

278.  Wetherall NT, Trivedi T, Zeller J, et al. Evaluation of neuraminidase enzyme assays using different substrates to measure susceptibility of influenza virus clinical isolates to neuraminidase inhibitors: report of the neuraminidase inhibitor susceptibility network. J Clin Microbiol2003;41:742–750.

279.  Pizzorno A, Bouhy X, Abed Y, et al. Generation and characterization of recombinant pandemic influenza A (H1N1) viruses resistant to neuraminidase inhibitors. J Infect Dis 2011;203:25–31.

280.  Okomo-Adhiambo M, Sheu TG, Gubareva LV. Assays for monitoring susceptibility of influenza viruses to neuraminidase inhibitors. Influenza Other Respir Viruses 2013;7(Suppl 1):44–49.

281.  Dharan NJ, Gubareva LV, Meyer JJ, et al. Infections with osltamivir-resistant influenza A (H1N1) in the United States. JAMA 2009;301:1034–1041.

282.  Hsu J, Santesso N, Mustafa R, et al. Antivirals for treatment of influenza: a systematic review and meta-analysis of observational studies. Ann Intern Med 2012;156:512–524.

283.  McNicholl IR, McNicholl JJ. Neuraminidase inhibitors: zanamivir and oseltamivir. Ann Pharmacother 2001;35:57–70.

284.  Gubareva LV, Kaiser L, Hayden FG. Influenza virus neuraminidase inhibitors. Lancet 2000;355:827–35.

285.  Hayden FG. Newer influenza antivirals, biotherapeutics and combinations. Influenza Other Resp Viruses 2013;7(Suppl 1):63–75.

286.  Stiver G. The treatment of influenza with antiviral drugs. CMAJ 2003;168:49–56.

287.  Shigeta S. Targets of anti-influenza chemotherapy other than neuraminidase and proton pump. Antivir Chem Chemother 2001;12(Supp 1):179–88.

288.  Hayden FG, Cote KM, Douglas GJR. Plaque inhibition assay for drug susceptibility testing of influenza virus. Antimicrob Agents Chemother 1980;17:865–870.

289.  Belshe RB, Smith MH, Hall CB, et al. Genetic basis of resistance to rimantidine emerging during treatment of influenza virus infection. J Virol 1988;62:1508–1512.

290.  Valette M, Allard JP, Aymard M, et al. Susceptibilities to rimantadine of influenza A/H1N1 and A/H3N2 viruses isolated during the epidemics of 1988 to 1989 and 1989 to 1990. Antimicrob Agents Chemother 1993;37:2239–2240.

291.  Potier M, Mameli L, Bélisle M, et al. Fluorometric assay of neuraminidase with a sodium (4-methylumbelliferyl-D-N-acetylneuraminate) substrate. Anal Biochem 1979;94:287–296.

292.  Buxton RC, Edwards B, Juo RR, et al. Development of a sensitive chemiluminescent neuraminidase assay for the detection of influenza virus susceptibility to zanamivir. Anal Biochem 2000;280:291–300.

293.  Chonel JJ, Pardon D, Thouvenot D, et al. Comparison between three rapid methods for direct diagnosis of influenza and the conventional isolation procedure. Biologicals 1991;19:287–292.

294.  Deyde VM, Okomo-Adhiambo M, Sheu TG, et al. Pyrosequencing as a tool to detect molecular markers of resistance to neuraminidase inhibitors in seasonal influenza A viruses. Antivir Res 2009;81:16–24.

295.  Moscona A. Global transmission of oseltamivir resistant influenza. N Engl J Med 2009;360:953–956.