Use of HIV Resistance Testing in Antiretroviral Therapy Decision Making
As HIV replicates, mutations in the HIV genome develop due to errors in the transcription of RNA to DNA by the viral enzyme reverse transcriptase. When these errors are introduced into viral genes, a mutation may result. If the mutation occurs in one of the HIV proteins that is a target of an antiretroviral drug, the result may be decreased susceptibility or resistance to that drug, and lack of inhibition of viral replication by that drug. All progeny virions that are produced from a cell harboring mutant, resistant virus contain the same mutation or set of mutations. Approximately one mutation is introduced into the virus genome with each cycle of virus replication.1 Because HIV replicates at such a rapid rate -- roughly one to 10 billion viral particles are produced daily2 -- virtually all possible mutations in the HIV genome are generated within a patient on a daily basis. In this way, all HIV patients, including those naive to therapy, harbor a diverse population of viruses with differing susceptibilities to the currently available antiretroviral drugs. When a patient starts antiretroviral therapy, failure to achieve or maintain plasma viral loads below quantifiable levels invariably leads to the selection of virus that contains mutation(s) that confer a survival advantage to the virus; in this case, there is some degree of resistance to one or more drugs within the patient's combination. Virus that is resistant to a drug within one class of antiretroviral agents is often cross-resistant to other drugs within the same class. Thus, when patients develop resistant virus, the potential to construct effective combinations of antiretroviral medications declines quickly. In order to achieve the goal of inhibiting virus replication and maintaining immunologic function in individuals who will live with HIV infection for decades, the selection of combinations that limit resistance and maintain therapeutic options for those who fail is essential.
Assays are now available that allow for the identification of resistant virus. The value of resistance assays has been validated by (1) improved outcomes in randomized clinical trials in which treatment decisions are made with resistance data compared to those made without this information and (2) from clinical trials that demonstrate improved virologic outcomes when patients receive more agents to which their virus is sensitive as determined by resistance tests.3-5
Types of Resistance Assays
Assays that report HIV resistance do so in two ways. Phenotypic resistance tests measure the concentration of drug needed to inhibit the replication of a patient's virus. Typically, this is quantitated by specifying the concentration of drug needed to inhibit 50% or 90% of virus replication (IC50 or IC90), or by comparing the fold-change in drug concentration required to inhibit the replication of the patient's virus compared to a representative, wild type, sensitive virus isolate. Genotypic resistance tests report the presence of specific mutations in the amino acid sequences of the HIV genome that encode the reverse transcriptase or protease enzymes, the targets of the HIV reverse transcriptase and protease inhibitors, or the part of the HIV genome that encodes a specific region that is the target of the HIV entry inhibitor.
Interpretation of Resistance Tests
For phenotypic tests in order to know whether a drug is potentially active against a virus, one must compare the IC50 or fold-change in virus susceptibility to a particular drug to the "clinical cut-off" of that drug. The clinical cut-off refers to the fold-change of virus susceptibility above which the drug has less activity in vivo. Often there are two cut-offs. Virus with a fold-change in susceptibility below the lower cut-off is fully susceptible, while virus with a fold-change in susceptibility above the higher cut-off is very unlikely to be inhibited at all. Virus with a fold-change between the cut-offs is partially susceptible. One of the limitations of the use of phenotypic resistance tests is that clinical cut-offs have not been clearly established for some of the nucleotide reverse transcriptase inhibitors or protease inhibitors used alone or boosted with ritonavir.
Genotypic resistance tests report the codon(s) in the amino acid sequence of the virus that differs from wild type. Only those mutations with known impact on virus susceptibility are typically listed. With each mutation reported three pieces of information are usually given: the number of the codon in the amino acid sequence that is mutated, the wild-type base, and the mutant base. Genotypic resistance tests provide an interpretation that assesses the impact of the particular set of mutations observed in the patient's virus on the susceptibility of that virus to available drugs. This interpretation is derived in one of two ways. One method is the application of an algorithm based upon a set of rules that link specific mutations with known patterns of resistance to a drug. The algorithms used for interpretation need to be regularly updated in order to include newly described mutations associated with resistance. Rules-based algorithms often fail to consider the interaction that several mutations may have on virus susceptibility. The second method for interpretation of genotypic resistance is the VirtualPhenotypeTM (Virco). In this system, the sequence of the patient's virus is matched with viruses that have similar genotypes stored in a database. The viruses represented in the database have had phenotypic virus susceptibilities performed. The virtual phenotype provides the average fold-change in IC50 of these viruses, gives the approximate proportions of matched viruses in the database that are fully or partially susceptible or resistant, and indicates whether the patient's virus is more likely to be sensitive or resistant. The virtual phenotype does not report whether the patient's virus is sensitive or resistance; it provides an estimate of the probability that the virus is sensitive or resistant.
Limitations of Resistance Testing and Discordance Between Phenotypic and Genotypic Resistance Test Results
There are certain limitations to the use of resistance tests. Because the current phenotypic and genotypic resistance test methodologies require PCR amplification of segments of the HIV genome, these assays may not be successfully performed when the patient's viral load is low. Generally, the viral load needs to be above 500 to 1,000 copies HIV-1 RNA/mL to obtain a result. In addition, resistance tests should be obtained with the patient continuing on therapy. Resistant virus in patients who stop therapy may decline in concentration as it is out-competed by wild type virus that is warehoused within latently infected cells.6 The resistant virus will re-emerge if the selective pressure of therapy is resumed. Therefore, if patients have been off therapy for one or more months, it may be best to resume therapy for a period of time prior to obtaining a resistance test. Resistance tests results are, most reliably, a reflection of the pattern of resistance to the drugs the patient is currently taking. Mutations present at one time point may not be detected at a second time point after a patient has switched therapy. Mutant virus on one combination that is lost following a change in therapy will reemerge if the patient cycles back to drugs that were used previously. For these reasons, when considering modification of an antiretroviral combination, resistance tests are a better indication of what drugs will not be effective, rather than an indication of what drugs will be effective. In addition, knowledge of prior patterns of resistance may be of value when selecting a new combination with the knowledge of a recent resistant test result.
Indications for Use of Resistance Tests
The International AIDS Society (IAS) -- USA and the Panel on Clinical Practices for Treatment of HIV Infection of the U.S. Department of Health and Human Services (DHHS) have published recommendations on the use of resistance tests.7,8 Both expert panels recommend resistance testing in patients with acute HIV infection or recent HIV infection, defined as seroconversion within the past one to two years. This recommendation is based upon a study that identified a cohort of patients with acute HIV infection, and demonstrated that of those who were infected in 1999 and 2000, 14% were infected with drug resistance virus.9 In addition, drug resistance can be quite stable in this group of acutely infected individuals, with resistance to all classes of agents persisting for over a year in some patients.10 The guidelines differ with respect to recommending resistance testing prior to initiating therapy in patients with chronic infection of more than two years duration -- the IAS panel recommends testing, while the DHHS panel recommends considering it. A recent report in 10 major U.S. cities demonstrated that patients with chronic HIV infection were just as likely to harbor resistant virus as those with acute infection.11 Resistance to the nucleoside reverse transcriptase inhibitors was most common. Therefore, resistance testing in antiretroviral naive patients is increasingly common, irrespective of the duration of infection. Resistance testing is recommended in pregnant women with quantifiable viral loads in order to optimize therapy and minimize the risk of vertical transmission.
Resistance testing is recommended in all patients with virologic failure prior to beginning a new antiretroviral combination. Patients with a quantifiable viral load on their initial combination may have virus that is resistant to only one, or perhaps two, agents in their combination. These individuals have many options for their next combination, and their therapy should ideally be switched to three new drugs, even if they exhibit virus that is only resistant to a single agent. In patients who fail multiple regimens, the pattern of resistance is typically complex. Resistance testing is used to optimize the therapeutic response, but the goal of achieving an undetectable viral load may not be possible. In addition, if a patient's CD4+ lymphocyte count is high, it may be prudent to withhold an agent or class of agents if it is unlikely that the patient will achieve an undetectable viral load, in the hopes that when newer agents are available a combination can be constructed that will be more successful in reducing the viral load below detectable levels.
Selecting Genotypic or Phenotypic Resistance Testing
Phenotypic resistance testing is more costly than genotypic testing. Therefore, genotyping may be preferred in resource limited settings. However, in true resource limited areas, neither of these may be an option. Despite the high cost of this monitoring, the selection of the most effective combination, and the prevention of additional virologic failure, virus resistance, and CD4+ count decline will be the most cost effective strategy in the long run, as it better achieves the most effective combination of drugs and limits or delays treatment failure. A genotypic resistance test is usually adequate when testing a treatment naive patient prior to initiating therapy or evaluating a patient failing on their first combination. However, the more complex the mutational pattern, the greater the value of a phenotypic resistance test. Patients who have failed more than three combinations will often harbor multiple mutations, and both a genotypic and phenotypic resistance test may be necessary to optimize the next combination.
Dr. Ian Frank is an Attending Physician; Director, Antiretroviral Clinical Research; and Associate Professor, Department of Medicine, University of Pennsylvania. Disclosures: Speaker's Bureau for both Virco and Virologic.
Transcript of The Body's Live Chat With Nelson Vergel on Issues for HIVers With Multi-Drug Resistance
This article was provided by Brown Medical School. It is a part of the publication Infectious Diseases in Corrections Report.