The Positive Side

Fall/Winter 2005 

Resistance Assistance

The ins and outs of HIV drug resistance testing

By Carol Major

THE TROUBLE WITH HIV is that it’s so stupid… it’s smart! Owing to its own “design flaws,” the virus sometimes manages to outwit the drugs used to keep it under control. This has led researchers to develop sophisticated scientific methods for analyzing drug resistance, helping us to keep up with some of the treatment challenges that HIV can throw our way.

The HIV virus is a mindless little robot, programmed to do just one thing: make millions more copies of itself. Antiretrovirals, the drugs that fight HIV infection, are designed to interfere with this replication process and prevent the tide of new virus from being formed.

If HIV always made perfect new copies of itself, these drugs would actually have a much easier job. An antiretroviral drug would face the exact same task every time, so if one were effective, it would always be effective. Unfortunately, nature doesn’t work that way: HIV, being a little bit “stupid,” makes mistakes every time it builds a new virus. These mistakes, or mutations, lead to slight differences in every new virus. Sometimes these new, mutated viruses can resist the drugs that should be controlling them… making it difficult to find a drug combination that keeps working.

Resistance testing is a way of finding out which drugs are still likely to be effective against mutated strains of virus — and which ones will not work. Resistance testing is a rather complicated business, even for experts. Your inner science geek may be curious about how it works — it’s actually quite the story. More pragmatically, you should know when it’s important to have testing done and what the results can (and can’t) tell you.

HIV’s gene machine

Like every living thing on earth, HIV is based on genetics. HIV’s genes — long string-shaped molecules — spell out the entire instruction book for building new viruses.

HIV has a natural tendency to mutate. That means that its genes and therefore the virus itself are slowly but constantly changing. Genetic changes lead to mutations — corresponding changes in the virus’ structure, shape and response to drugs.

By studying these mutations, investigators can potentially map different subtypes of HIV in different regions of the world and determine whether one particular person may have infected another. However, the most important use of HIV genetics is identifying drug resistance.

Where does drug resistance come from?

Mutations of some sort are produced every time the virus replicates. This usually happens more or less at random — purely as a result of mistakes in the replication process. Many of the resulting mutations don’t make much of a difference:

  • Silent mutations change the genetic makeup slightly but do not physically change the virus in any detectable way.
  • Polymorphisms change the physical makeup of the virus but do not necessarily cause drug resistance. In fact, some of these polymorphisms actually harm the virus: These “crippled” variants can’t survive or reproduce as well, so they tend to die out quickly. Others are fairly neutral — the new virus is different but not in any significant way. (Similarly, human genetic differences result in different eye or hair colour, or whether or not you can curl your tongue.)

Even with these mutations, some viral strains — called wild-type — tend to predominate. Despite the name, these “wild types” aren’t microscopic party monsters; they’re just the “naturally occurring” forms of HIV — the strains that tend to persist in people who are not taking anti-HIV medications.

Then there are other, nastier forms of mutations:

  • Some mutations cause drug resistance by physically changing the virus so that it is no longer sensitive to one or more drugs. (Drug sensitivity is a measure of how well a drug suppresses, or controls, the virus.) Some of these are classified as primary mutations — the “prime suspects” that severely limit a drug’s effectiveness.
  • Any mutation that doesn’t completely cripple the virus will be passed on to its “offspring,” which can then accumulate more. Therefore, after a few generations, a single virus may have many mutations. Some of these so-called secondary mutations do not cause drug resistance on their own but do contribute en masse if enough specific ones pile up.
  • Some mutations do not affect the amount of drug resistance a virus has but, instead, help a virus with resistance mutations to grow more quickly. These are called compensatory mutations.
  • Finally, some mutations actually reverse the effects of others, so that a virus with a resistance mutation may become somewhat or completely sensitive to that drug again. These are called antagonistic mutations.

How do mutations produce drug resistance?

HIV needs to produce an assortment of proteins, like a viral “toolkit,” in order to do its job. Antiretroviral drugs generally work by a lock-and-key effect that prevents one or more of these proteins from doing its job. However, if a viral protein physically changes in just the right way, that mutation may allow the virus to escape from the drug’s “locking” effect. Drug-resistant mutations cause these specific changes to viral proteins — changes that usually weaken the virus slightly but allow it to evade the drugs.

HIV doesn’t necessarily want to mutate into a drug-resistant form. However, when the virus is not completely suppressed, billions of new and different viruses are continually being produced. It can be just a matter of time before the virus stumbles on a mutation that gives it some resistance to a particular drug. (Picture an obsessed locksmith making keys at random. Sooner or later, one may fit the right lock.) If you aren’t on meds, drug-resistant viruses may appear by chance, but they don’t have any particular advantage. In fact, they are usually weaker than their unmutated relatives and don’t reproduce fast enough to survive very long.

However, if you are on meds, the viruses are swimming in a sea of antiretroviral drugs that are trying to stop them from replicating. If one little virus has a mutation that allows it to replicate easily in the presence of those drugs, even while all its relatives remain suppressed, the resistant virus will “preferentially” replicate until it becomes the main virus in the mix.

Also, since drugs within each class are structurally similar, resistance to one may mean resistance to others in its class as well — a scenario known as cross-resistance.

But I take all my meds — why do I have resistant virus?

Resistance mutations are most likely to arise in the presence of less-than-effective drug concentrations in the blood. Scientists have shown that patients taking some but not all of their meds have the fastest development of mutations. That’s why there is so much emphasis on adherence. It is important to try to have all of your drugs in your system at all times to keep resistance from emerging. Most importantly, do not take only one or two drugs of your combination.

Low drug concentrations can also occur in people who are fully adherent to their medication, due to factors such as poor absorption, rapid drug clearance and drug interactions. Therapeutic drug monitoring, which measures drug concentrations in the blood, might be helpful for identifying these problems, but it is generally not available. Drug interactions, however, can be avoided with careful prescribing.

Detecting drug resistance

There are two main methods for detecting drug resistance: phenotyping and genotyping.

The seemingly most direct way is to try to grow the virus in a test-tube in the presence of measured amounts of drug. If it grows, it is resistant to the drug; if it does not grow, it is sensitive. This is called phenotyping. Phenotyping measures just how sensitive or resistant a virus is — whether it grows quickly, slowly, or not at all. Though phenotyping seems like the most straightforward method, it is technically quite difficult to do and is rarely used in practice. Phenotyping is not routinely done in any labs in Canada, although Dr. Mark Wainberg has recently arranged a pilot project in Quebec, sending specimens to the Virologic laboratory in California.

An alternative method, called genotyping, is by far the most common method of assessing HIV drug resistance. Although the process seems (and is) very complicated, it boils down to a job of translation — somewhat like reading Egyptian hieroglyphics or computer code. At the resistance testing labs, technicians take a “viral fingerprint” (the genetic structure of the virus) and turn it into tangible information: how well various drugs can control HIV. When we look for mutations, we are comparing the individual viral fingerprint to the normal wild-type virus. Any differences from wild-type represent spots where the virus has mutated.

The final, crucial step in genotyping is to assess whether these mutations represent drug resistance. There are three major commercial genotyping technologies that do this in different ways:

  • vircoTYPE uses a “virtual phenotype.” This is a large and growing “library” that matches up known mutations with known degrees of resistance to various drugs.
  • Bayer TRUGENE uses an expert panel to assess and identify the specific drug resistance mutations.
  • ViroSeq matches resistance mutations to a standard database to determine the potential for drug resistance.

Though all three technologies provide similar results, there are important differences. For example, the virtual phenotype gives an estimate of the degree of resistance caused by a given set of mutations, while the other approaches only state whether you are resistant or not. In addition, sometimes the approaches used to interpret the mutation data can result in entirely different results for the same sample.

Genotyping for assessing HIV drug resistance is widely available in Canada. Different provinces have different guidelines and use different kits or methods, but usually physicians can order the test by indicating a viral load rebound or failure to respond to therapy. Genotyping immediately after seroconversion can be more problematic but can usually be arranged. It is also becoming increasingly important to have phenotyping availability to sort out treatment options for those with complex drug resistance patterns.

What does genotyping tell you?

Resistance test reports generally provide a detailed list of all of the specific mutations found in the viral population tested, plus a summary of the corresponding levels of drug resistance. The summary often uses colour codes to flag varying degrees of drug resistance. The detailed mutation data allow physicians to combine their own expertise with the summary and interpretations of the report.

Certain cautions have to be observed when interpreting a genotypic test result. These include the following:

  • Viral load must be more than 250 copies/ml for the test to be successful.
  • When assessing treatment failure, patients should be actively taking medication for the test results to be accurate. Mutations cannot be detected during treatment interruptions even though they may still be around.
  • Mixtures of mutations (more than one mutation at the same “site”) sometimes may not be detected by the test if they are present in small amounts.
  • Cumulative records of drug resistance tests should be reviewed when making treatment decisions.

Drug resistance testing is just one of many factors to be considered when making treatment decisions — along with your treatment history and your own judgment as to how you can handle side effects and the demands of drug adherence.

When to test

There are two points in time when it is useful to do drug resistance testing.

  1. Drug resistance test results are particularly important if and when drug therapy fails. If you’ve been on therapy for some time and your viral load begins to “break through” (rise again), a prompt resistance test can identify which mutations have occurred. This information is invaluable for choosing a successful next regimen before more mutations accumulate and further complicate the picture.
  2. A test should be done as soon as possible after diagnosis. If you are infected with an already-resistant strain of HIV but are not actively taking meds, the mutated HIV strain will diminish until it is not detectable by resistance tests. The transmitted mutations are still stored in your body, even though they may seem to disappear as the majority of the virus naturally “reverts” to wild-type. A resistance test done later might not pick up these mutations and may not provide an accurate picture of all the drugs you could be resistant to.

Dr. Paul Sax of Boston, Massachusetts, recently published a paper showing that resistance testing at initial HIV diagnosis — before treatment is started — can guide a doctor to select the appropriate treatment regimen and improve a patient’s outcomes. In Canada, between 7% and 15% of those just diagnosed have HIV resistance mutations. So, even if you have no intention of starting therapy soon, a resistance test done as soon as possible after infection provides valuable information for future medication decisions. A more cost-effective approach might be to store the sample and do the test when you decide to start treatment.

Acting on results

Doctors and PHAs face many challenges in balancing the results of these tests with quality of life and other factors. For instance, there is still somewhat of a dilemma in deciding how to handle an otherwise-healthy PHA who has a low but detectable viral load (say, in the low thousands) on a well-tolerated regimen. Sometimes the push to undetectable has very undesirable effects related to tolerance and toxicity. On the other hand, switching may preserve the usefulness of the other drugs in the regimen. With pros and cons on both sides, this is a decision to be made carefully by the doctor and patient.

One of the newest directions in resistance testing is the concept of the “virtual virus.” Once a mutation occurs, it is “archived” in your blood cells. It may then be “overgrown” by other viral strains, so that subsequent resistance tests may very well miss it. However, the archived mutation remains ready to re-emerge if the drug that caused it appears on the scene again.

For this reason, Dr. Julio Montaner and colleagues at the B.C. Centre for Excellence in HIV/AIDS have realized that it is critical to track the virus over time, taking into account the accumulation of mutations even though they may not be present in the most recent sample. (On the other hand, a virus with many mutations may be slightly less virulent, or damaging, than a wild-type virus because a virus with mutations may be slightly less “fit” — the virus pays a price in its ability to make copies of itself when it picks up mutations.)

New drugs for treatment are continually being developed. We know that it is likely that these new drugs will give rise to new resistance mutations. Strategies for treatment need to keep the viral load completely suppressed for as long as possible. But the good news is that this is already the norm. The majority of PHAs now stay suppressed for years without changing meds.

Carol Major is a consultant at the Ontario HIV Treatment Network and former head of the HIV Laboratories for the Ontario Ministry of Health.

Retro science: DNA, RNA and all that jazz

HIV is from the family of retroviruses. Although these days being “retro” is akin to being cool, in the case of HIV it literally means backwards. Whereas the HIV virus’ genetic material is RNA (ribonucleic acid), most of the rest of nature mainly uses DNA (deoxyribonucleic acid) as genetic material. RNA is used primarily for the grunt work — sending messages, shunting building blocks back and forth, and other construction activities.

HIV, being retro, takes the unusual step of using RNA for its genetic material. It then has to work backwards, making DNA from its RNA before it can actually reproduce itself. This requires an enzyme called reverse transcriptase.

Reverse transcriptase doesn’t do this very well at all. Rather than creating exact replicas of new DNA, it continually makes mistakes. This is where mutant viruses come from.

Basic steps of genotyping

  1. Pick apart the viral genes to isolate the ones that deal with how the virus reacts to drugs.
  2. “Amplify” these genes to make sure there is enough for the test to work on.
  3. Lay out the exact chemical structure of the genes in a process known as genetic sequencing.
  4. Analyze this chemical sequence, or genetic “fingerprint,” to determine how much the virus is resistant to various drugs.

All genes — whether they’re yours, a housefly’s or HIV’s — are made of long strings of just four smaller molecules (“nucleotides”) repeated in a very precise coded order. Think of a long string of beads — black, white, red and blue. That long sequence of coloured beads contains all the information needed to build a new virus.

Mutations are places where the “normal” colour has been replaced by a different one — a blue bead, say, where a red one should be. (In fact, the “coloured beads” are called nucleic acids, and we represent them with the codes A, T, C and G. So a string of normal viral DNA might be spelled “ATCG,” and a mutated variation might be “ATTG.” Computers are perfectly suited to scan the seemingly endless strings of letters for differences — a task no human would want to tackle.)

Reading resistance

  1. Mutations are the exact locations on the HIV gene that have changed from “wild type.”
  2. Amount of resistance / loss of drug effectiveness (higher numbers = higher resistance = less effective drug)
  3. This is the bottom line: how effective the drug will be. (Maximal = most effective; Susceptible = good; Reduced response = less effective; Resistant or Minimal = not effective)