HIV reproduces at an incredible rate, producing more than 100 billion new copies of virus each day. As the virus copies its genetic information, there are random errors, or mutations, that occur. These mutations have different effects. Sometimes, mutations make copies of the virus that are less "fit" than the wild-type (original) virus and are unable to infect immune system cells. Other times, the mutations have no effect on the virus at all.

Occasionally, a mutation creates a copy of the virus that is resistant to antiretroviral medication. Different numbers of mutations are required to render different medications ineffective. For example, one viral mutation can lead to 3TC resistance, while several mutations are required to develop resistance to most protease inhibitors. Resistant virus cannot be controlled as effectively with antiretroviral medication as wild-type virus, and viral load increases as a result. This is called treatment failure. The virus is not being smart or tricky when a resistance mutation occurs; the mutations are merely a result of rapid copying.

Resistance to an entire class of drugs, not just one drug, is typical. Someone who has taken indinavir (Crixivan, a protease inhibitor), for example, may be resistant to all protease inhibitors. This situation is known as cross-resistance.

What’s the relationship between undetectable viral load and resistance? If viral load is undetectable, the amount of viral reproduction is low, and the number of mutations generated is low. This results in fewer strains of resistant virus.

Let’s take a step back, and talk about antiretroviral therapy in general. The goal of antiretroviral therapy is to reduce the amount of HIV in the blood to a level that is as low as possible for as long as possible. The best way to achieve an undetectable viral load is with highly active antiretroviral therapy (HAART), a combination regimen containing medications that prevent viral reproduction at various stages of the process.

An initial regimen that brings the viral load to undetectable is the best chance at long-term suppression of HIV. When taken in the prescribed manner, an initial treatment regimen produces undetectable viral load in about 80% of patients. Salvage regimens taken after the initial regimen have a lower success rate, with about 30% of patients returning to an undetectable viral load.

Development of resistance can lead to treatment failure. Other factors, like non-adherence to an initial regimen, incomplete viral suppression, a high baseline viral load, poor penetration of the drug into tissues like the lymph nodes where the virus lingers, and pharmacokinetic factors (how well or poorly the drug is absorbed and processed) can also lead to treatment failure.

With these factors in mind, Dr. McGowan recommends starting a salvage regimen if:

• Viral load has never reached undetectable;
• Viral load increases after having been undetectable;
• T-cell count decreases significantly; or
• There has been clinical deterioration.

Genotypic testing identifies specific viral mutations, some of which are associated with a particular medication or class of medications. Phenotypic testing is a direct measure of the effectiveness of various drugs against an individual’s current prominent strain of virus. Other strains may be present, but not in large enough amounts to be examined by the test. The other strains may have mutations that were developed during treatment with previous medications. Therefore, phenotypic test results must be used in conjunction with medication history when used to develop a salvage regimen, as they only provide information about resistance in the current regimen.

Q: Please comment on "drug holidays."

A (Gormley): The 3-month drug holiday that I took before I started my current regimen was recommended by my doctor. It made me a little nervous to be off all of my medications, and I was ready to begin treatment again at the end of the holiday.

A (McGowan): If the T-cell level stays high and the viral load stays low without medications, there may be no detriment to taking a drug holiday. If ceasing medications makes the viral load increase, and the T-cell count drops, it may be difficult to regain these T-cells.

If drugs are stopped because of toxicity, it may be possible to substitute one component of the previous regimen, or to try a different regimen entirely. Just remember to stop all drugs at the same time when stopping medication. Stopping just one drug could lead to resistance.

Q: Can you "recycle" drugs in salvage therapy?

A (Gorman): I am currently on ddI, after having tried it twice. If you run out of new therapies to use, you may have to use a drug again, and may be able to tolerate it the second (or third) time around better than you could before.

A (McGowan): It may be possible to "recycle" drugs, that is, to use drugs in a salvage regimen that have been used before. If underlying resistance is present, however, the drug will become ineffective more quickly that it would have had the treatment never been used. Success rates in salvage therapy are highest when a drug from a new class of drugs is used.

Q: What happens when someone is infected with a strain of virus that is already resistant to medications? Should everyone have a phenotypic test before they start their initial treatment regimen?

A (McGowan): Infection with a strain that is already resistant to some anti-HIV medications is called "primary resistance." This has been of increasing concern lately, especially in an urban area like NYC where many individuals have access to treatment. The problem with performing phenotypic testing is that unless it is performed immediately after someone is infected, the virus may have already reverted, through mutations, to wild-type virus. The phenotypic testing will not capture the drug-resistant strain. The only way to obtain this information would be by learning the treatment history of the person who transmitted the virus, which, for obvious reasons, could be complicated.

Q: When is it most appropriate to use phenotypic testing? Will my insurance company pay for it?

There are several situations in which the testing is merited:

• When failing a regimen, to identify medicines that might not be effective.
• To rule out primary resistance, if the test is performed soon after infection.
• In pregnancy, to determine which combination will be most effective in preventing transmission from mother to fetus.

Insurance companies do not generally cover the cost of the test because it is experimental (has not been approved by the FDA).