Resistance Testing

July 2006; updated July 2007

Chapter Contents

Background

Modifying Factors

Key Points

References

Table 1. Resistance Testing Recommendations

Background

As of mid 2006, 2 major types of resistance testing are available: genotype tests and phenotype tests.

Genotype Tests

Genotypic testing works by amplifying and sequencing HIV taken from a patient to look for mutations in the HIV reverse transcriptase and HIV protease genes, which are known to correlate with clinical resistance to antiretroviral drugs. This test generally can detect mutations in plasma samples with HIV RNA levels >1,000 copies/mL. Species representing 20% or more of the amplified product usually can be detected by current techniques. Minor species may not be detected. Resistance mutations acquired in the past, under the selective pressure of a previous drug, may be archived in minor species and remain invisible to genotypic testing. These resistance mutations may reemerge and cause drug failure, however, if the previous drug is used again.

A genotype test takes 1-2 weeks to complete. The results are reported as a list of the mutations detected; most reports also include an interpretation that indicates the drug resistance likely to be conferred by those mutations (see "Modifying Factors" below, for a discussion of the limitations of resistance testing).

Genotype results can be difficult to interpret. A thorough antiretroviral history and expert clinical review, therefore, are necessary to put the results of a genotype test in proper perspective and to identify options for further treatment. A compilation of the most common HIV mutations selected by the 3 classes of antiretroviral agents is available at: http://hiv-web.lanl.gov .

A "virtual phenotype" is a genotype that is compared with a databank of patients' samples that have been analyzed by paired genotype and phenotype testing. The patient's genotype is matched to a banked genotype, and the patient's phenotype is then predicted based on the phenotypes paired to the banked genotype. A virtual phenotype can be completed in the same amount of time as a genotype. Results are reported as a genotype (listing the mutations detected) as well as a predicted fold change in the 50% inhibitory concentration (IC50) of each drug to the patient's virus (see "Phenotype Tests" below). The predicted susceptibility of the patient's virus to each drug is then reported, based on biologic and clinical cutoffs.

Phenotype Tests

Phenotypic testing works by splicing the HIV reverse transcriptase and HIV protease genes from a patient's virus into a standardized laboratory strain, which is then grown in the presence of escalating concentrations of antiretroviral drugs. The test measures the IC50 of each drug against the virus in vitro. Results are reported as fold-change in IC50, as compared with a drug-susceptible control strain or with a previous test of the same patient's blood. The predicted susceptibility of the patient's virus to each drug is then reported, based on what is known about the correlation between fold-change in IC50 of that drug and clinical resistance. As with genotypic testing, the phenotype may not be able to detect resistance if the HIV RNA is low (<1,000 copies/mL) and may not detect minor species. A thorough antiretroviral history and expert interpretation are essential in determining the significance of the results. A phenotype takes 2-3 weeks to complete.

Modifying Factors

Table 1 presents an overview of when genotype and phenotype testing is, and is not, recommended.

Limits of Resistance Testing

In a patient taking antiretroviral therapy (ART), drug-resistant HIV evolves in response to selective pressure applied by the antiretroviral drugs in the patient's system. Specific resistance mutations develop in response to the pressure exerted by specific drugs (M184V, for example, evolves in response to lamivudine or emtricitabine). The presence of viral resistance suggests that a particular drug (and drugs with similar resistance patterns, or cross-resistance) is unlikely to be successful in suppressing viral replication.

In contrast, the absence of resistance to a drug does not necessarily indicate that the drug will be successful, particularly if that drug (or drugs sharing cross-resistance) has been used previously. If a particular drug is discontinued, the viral strains harboring the mutations that confer resistance to that drug may decrease below the threshold of detection by the resistance assay, so the resistance test may not reveal certain resistance mutations. In such situations, minority populations of resistant viruses may exist in reservoirs and may emerge rapidly under selective pressure if that drug is restarted, or if drugs with similar or overlapping resistance patterns are used. The implications of archived mutations are 2-fold: 1) Resistance tests are most reliable while the patient is still on the failing regimen; and 2) resistance testing should be interpreted in the context of both the drugs that the patient was taking at the time of the test and the drugs that the patient had been exposed to previously (ie, the patient's antiretroviral history).

Antiretroviral-Naive Patients

In treatment-experienced patients, as indicated above, resistance testing is most reliable when performed while the patients are still taking the failing antiretroviral medications. In treatment-naive patients, resistance testing may reveal resistance mutations that were acquired at the time of infection, through infection with a strain of HIV that had already developed antiretroviral resistance. Current guidelines recommend genotypic testing in recently infected patients and in antiretroviral-naive, chronically infected patients before initiation of therapy. Many experts suggest testing as early as possible in the course of HIV infection, to increase the likelihood of detecting transmitted mutations. The rationale for resistance testing in antiretroviral-naive patients is 2-fold: 1) The incidence of primary resistance is rising, particularly in locations with a high prevalence of persons taking ART; and 2) unknowingly starting a patient on antiretroviral medications to which his or her virus is already resistant may risk failure of the initial regimen, rapid acquisition of additional resistance mutations, and curtailment of future treatment options.

Using Genotype and Phenotype Tests at the Same Time

Genotype and phenotype tests have a few complementary properties that may, in some circumstances, make it desirable to use both tests at the same time. This strategy is especially advantageous when trying to devise a regimen for patients who have been exposed to many antiretroviral agents and have few remaining treatment options, and for whom the development of additional resistance could be particularly dangerous. For example, early mutations may appear on a genotype before detectable increases in inhibitory concentrations, and these would not be detected on a phenotype. Phenotypic testing can detect loss or gain of drug efficacy caused by complex interactions of mutations that, by themselves, would not be predictive.

Resistance Testing in Patients with Virologic Failure

Factors other than nonadherence and resistance may cause failure of ART; these include drug-drug interactions and malabsorption. Therefore, before assuming that drug failure is due to resistance and ordering a resistance test, it is important to assess the causes of antiretroviral regimen failure. If resistance is still suspected after assessing all possible causes, resistance testing should be done while the patient is taking the failing regimen, for the reasons noted above.

Key Points

• There are 2 major types of resistance testing currently available: genotype and phenotype tests.
• In general, a patient's viral load must be at least 1,000 copies/mL for either test to be reliable.
• Both genotypic and phenotypic testing can detect resistance only if it exists in at least 20% of the viral species present in a patient (known as the dominant species). Minor species may harbor resistance that remains undetected by either test.
• Resistance tests are most reliable when performed while a patient is still taking a failing regimen, or within 4 weeks after stopping. Neither test predicts which drugs will be active in a particular patient, only drugs that are not likely to be active. Nevertheless, studies comparing the use of resistance testing to expert opinion alone have shown that resistance testing can improve virologic control of HIV. Most treatment guidelines recommend resistance testing in certain circumstances.

References

• Bartlett JG, Gallant JE. 2005-2006 Medical Management of HIV Infection . Baltimore: Johns Hopkins University Division of Infectious Diseases; 2005. Available online at hopkins-aids.edu/mmhiv/order.html.
• Shafer RW. Genotypic Testing for HIV-1 Drug Resistance . In: Peiperl L, Coffey S, Volberding PA, eds. HIV InSite Knowledge Base [textbook online]; San Francisco: UCSF Center for HIV Information; April 2004. Accessed February 7, 2006. Available online at hivinsite.ucsf.edu/InSite?page=KB.
• U.S. Department of Health and Human Services. Guidelines for the Use of Antiretroviral Agents in HIV-1-Infected Adults and Adolescents . October 10, 2006. Available online at aidsinfo.nih.gov/Guidelines/GuidelineDetail.aspx?GuidelineID=7. Accessed July 3, 2007.