There have been a number of excellent studies published which support the use of resistance testing to help providers choose the best possible options that are available. The DHHS (Department of Health and Human Services) has embraced this information and has added the use of resistance testing to its HIV guidelines.

Currently, the DHHS recommends that resistance testing be performed or considered in several situations:

1) Virologic failure during combination HIV antiretroviral (ARV) therapy.

2) When the viral load is not fully suppressed after ARV initiation.

3) In acute (initial) HIV infection if ARV treatment is going to be started.

4) In chronic (long-term) HIV infection if it is possible the patient was infected by someone with drug resistant HIV.

The reason for testing someone at baseline before they have ever even taken a single pill of HIV medicine is because resistant HIV can be transmitted between sexual partners, from mother to child, or between IV drug users who share needles. If the source person has virus that is resistant to one or more of the HIV medicines (either because they have been on a failing regimen or they themselves were infected with resistant virus), there is a reasonable chance this same resistant virus will be the strain that causes infection in the new “host” or patient.

Since resistance is carried in the genes of the virus, it is not lost during transmission; it is passed along to the new patient. It is useful to know if this has occurred because the single best chance for long-term control of HIV is with the first regimen. If a mutation is detected with the genotype, the medications that are affected by it can be avoided right from the start.

The reason for obtaining resistance testing after treatment failure is because in most cases failure is associated with the development of resistance mutations which then accumulate over time. The longer someone is left on a failing regimen, the more likely it is that the virus will “collect” more and more mutations that allow it to evade more of the HIV medicines—even medications you may not have ever taken.

Genotype tests are most useful to clinicians when there are only a few mutations present. This is because when there are only a few (early) mutations present, their effect on the virus is easier to predict.

Once there are multiple mutations present (such as after failing the second, third, or more regimens), they begin to have complicated and unpredictable effects on the virus and each other.

Sometimes two mutations “cancel each other out”; sometimes their effects are additive (one mutation alone isn’t bad, but two or three or more together lead to resistance).

Occasionally mutations occur that are beneficial (weakening the virus or making the virus more susceptible to a different medicine).

One of the limits of genotype testing is that it requires a detectable viral load. Ideally the test is used when the viral load is greater than 1,000 copies/ml, although results can sometimes be acquired with lower levels.

Some people who experience viral breakthrough (a detectable viral load) develop lower-level viremia with copies only in the hundreds. It is possible that resistance is occurring at this time but the genotype test is not sensitive enough to detect it.

Further, in order for the genotype to detect a mutation, at least 20% of the viral population must contain the change.

The use of phenotype testing is limited by a few factors. First, it is more expensive than genotyping. In resource limited settings (such as public health clinics or with uninsured patients), use of this test may be restricted to contain costs. Second, the results typically take several weeks to return (whereas genotype results often can be reported in about a week). It too requires a viral load of about 1,000 copies/ml and is less sensitive/useful with lower levels.

The advantage of phenotyping over genotyping usually occurs once a patient has failed more than one regimen. As mentioned before, once multiple genotypic mutations have accumulated, their effects on each other become more complicated and more difficult to predict using algorithms or computer models.

With a phenotype, there is very little guesswork about the interplay of the mutations because the test is simply looking to see whether or not the virus can grow in the presence of the medicine. The test doesn’t know or care which or how many mutations are present. In a perfect world, this test would be widely available, very cheap, give rapid results, and be used together with a genotype.

With a newly infected or newly diagnosed patient, among the myriad of baseline tests that are ordered, a genotype should be considered based on the guidelines. In population centers where there is a higher prevalence of resistance (such as large cities) or patients who are infected from a higher risk group (such as IV drug users who share needles), this sort of genotypic testing is recommended. In areas where HIV is much less prevalent or access to medications is limited (rural communities, developing nations, etc.) the use of baseline resistance testing may not be cost effective due to the lower baseline prevalence of intrinsic resistance in the general population. Once the result is available, it is carefully evaluated to determine the initial regimen that will have the highest likelihood of long-term success.

Recommendations for resistance testing during pregnancy are the same as for anyone else. They may have high CD4+ counts or low viral loads that would otherwise negate the need for immediate treatment. However, given the risk of perinatal transmission (the virus infecting the infant) during delivery, most providers will place the mother on antiretrovirals at some point prior to labor to minimize the risk of exposing and infecting the infant. Baseline genotype testing is particularly useful in this setting because the critical component of treatment before delivery is achieving an undetectable viral load. This is most likely to be achieved if the virus is treated with at least three medications to which it is completely susceptible.

For people on a regimen that is failing, resistance test results are most useful if the patient has the blood drawn while they are still actually taking the regimen and to which resistance is suspected to have developed. This is because the most “fit” virus is the wild type (naturally occurring virus). Without the selective pressure of the ARVs ongoing, both the resistant and wild type virus will replicate, but since the wild type is most fit, it will “out-populate” the resistant virus quickly, and the genotype test may be unable to detect the resistant strain.

The same principal applies to phenotypic testing. In practice, some providers will ask their patients to remain on the “failing” regimen while they await the results of the genotype (or phenotype) test. Others believe it is prudent to discontinue the failing regimen as early as possible to limit the chances of additional mutations accumulating while they await the return of the test results.

In clinical practice, I have found resistance testing to be an increasingly valuable tool in long-term management of my patients. Ideally all regimens would be ironclad and last for decades, but the reality is that not only are the medications imperfect, but it is extremely difficult for anyone to maintain perfect adherence to them over the long haul. Resistance is almost inevitable at some point in anyone’s treatment history and genotyping/phenotyping eliminates much of the guesswork and allows for greater precision in selecting the next regimen.

Jail setting

I work with a special population. My primary care site is the Cook County Jail in Chicago. It is one of the largest single-site correctional facilities in the United States. We house over 11,000 detainees on any given day with an HIV positive rate of 2.6% based on a serosurvey performed in 2001, a little over five times the prevalence rate for the U.S. as a whole. This equates to approximately 300 HIV-positive people behind our bars on any given day.

We are an exception to the traditional management rules used by most outpatient providers for several reasons.

1) We fall under the “resource limited” category, meaning that due to cost constraints we do not routinely offer the more expensive phenotype testing (there are rare exceptions).

2) As a “resource limited” site, we also do not offer genotypes at baseline for newly infected or newly diagnosed patients (unless they provide a very solid history of exposure from a known or high-risk resistant source).

3) Since we are a jail (not a prison) our population is highly transient. The average length of stay is only nine days, so it is uncommon for either a genotype or phenotype result to be available before the detainee is released or sent to prison. Thus, the decision to use this type of testing is often deferred until the patient presents to their primary clinic as an outpatient.

4) Pregnant women are routinely offered genotype testing if we know they will remain in our facility long enough to receive the result.

Aside from being a resource limited setting, the single biggest differentiating factor in determining the practical use of resistance testing in corrections is the disposition of the inmate. In jail settings, detainees are held until their day in court. This means they remain under the care of the corrections healthcare team until they are either released, bonded out, or convicted and sent to prison.

In Cook County Jail, the average length of stay is only nine days, and more than 75% of those incarcerated are gone within 30 days. For this reason, there are only occasional instances where either type of resistance testing is practical because the detainee will more often than not be gone before the results ever return from the lab.

In prisons, the inmates have definitive sentences with known release dates. Knowing the patient will remain at one location long enough to observe trends in viral load testing, testing for resistance, waiting for the results, and then intervening is far more practical. The impact of proper use of genotype and phenotype testing is far more substantial in a prison population.

I also work at a “Continuity of Care” clinic at the CORE Center in Chicago. At this site, I am able to practice more traditional HIV medicine since the patients who attend this clinic will return to me for multiple visits just like any other primary HIV treatment setting. Although the CORE Center is still within the boundaries of a “resource limited” setting, the rules for ordering genotype testing are somewhat more relaxed compared to the jail.

Routine baseline genotypes are generally not performed excepting for the above mentioned special circumstances. All first-line treatment failures are genotyped as early as possible if the viral load levels rise above 1,000 copies/ml. Due to cost restraints, we do not offer phenotyping for advanced level treatment failures. Instead, genotyping is performed and the often complicated results are interpreted by a select committee of resistance experts within our facility who then suggest treatment options.

Chad Zawitz, MD completed his Infectious Diseases Fellowship at Rush University Medical Center, Chicago in June 2004, and started as attending physician at Cermak Health Services (Cook County Jail) in July 2004. He has been indirectly involved in HIV care since 1995, and began caring directly for HIV-positive patients as a senior internal medicine resident at University of Pittsburgh Medical Center in 2001-2002. At Rush during his fellowship, like all Infectious Diseases fellows, he had his own HIV clinic patients at the CORE Center (a joint venture outpatient HIV hospital associated with Rush and Cook County Hospital) for the past two years prior to starting full-time at the jail.