The key point to emphasize is that HIV can only become resistant to a drug if it is actively replicating (reproducing itself). When HIV viral loads are reported to be "non-detectable," this is very misleading. What a "non-detectable" result really means is that the amount of HIV present in the blood is below the limits of the viral load test to detect HIV. Most clinics in the country use tests which can detect 500 (or more) copies of HIV, per ml of blood. Thus, "non-detectable" means that you have 499 (or less) copies of HIV, per ml of blood. These tests are called "sensitive" viral load tests. There are now several companies which have "ultra-sensitive" viral load tests available, which can detect 20 (or more) copies of HIV, per ml of blood. Thus, a "non-detectable" result from such a test would mean that you have 19 (or less) copies of HIV, per ml of blood.

More accurately, the test result should be reported "below the limits of quantification", not "non-detectable". This is not a trivial point, as will be discussed in this article, the chance of HIV developing drug resistance may be very different if your viral load is 490, versus 0 copies; yet, in both situations the current, most widely-used viral load tests would report the result to be "non-detectable." "You'd better get that drug when it's new, because it won't work as well once it's been around for a while." The factors leading to this truism of medicine will explain many of the issues involved in the recent reports that the new anti-HIV drugs control HIV replication for only about half of all HIV-infected persons.

Reports by researchers from the University of California at San Francisco presented at the ICAAC meeting in Toronto, in September, 1997, showed that 53% of persons at their HIV clinic did not achieve long-term suppression of HIV replication with combination therapy. This contrasts to the controlled trials of highly active antiretroviral therapy (HAART) which showed that triple drug combinations, including a protease inhibitor (PI), completely suppressed HIV replication for 1-2 years in up to 80-90% of persons studied. Why, then, do only half of persons treated in the "real world" achieve a similar result?

Factors Contributing to Treatment Failure

Trial Participation Selection

Clinical trials are very selective in which people they enroll into the trial. "Inclusion criteria" often exclude persons with significant liver, kidney, of blood disorders.

A concrete example of the impact of this selection process is that people with chronic liver disease, such as many HIV-infected injection drug users with chronic hepatitis C infection, often experience worsening of their liver disease when treated with protease inhibitors (PI). Had these people been included in the initial PI trials, the results would not have been so good.

Additionally, controlled trials usually enroll persons who are more compliant, more educated, and more motivated to seek out the latest, best treatments. Some of these factors are correlated with better treatment results, than those attained by the broad spectrum of all persons seen in a clinic.

Some trials exclude persons who have been treated with PI's or reverse transcriptase inhibitors in the past. This may be done, for example, by a protocol which offers a new protease inhibitor (PI) to people who have failed all other PI's, but also requires that the person be on two reverse transcriptase inhibitors (RTI) which they have never taken before. If a person has been on (and failed) all the RTI's currently available, then they will not be able to participate in that trial. The fact that a drug performs better in a controlled trial than in the "real world" is not unique to anti-HIV drugs, but it has created inflated expectations of the efficacy of the new anti-HIV drugs in the eye of the public, many health care providers, and people living with HIV.

Compliance

Due to the unique ability of HIV to mutate, drug resistance can arise when a person misses even a couple of doses of a drug (i.e. "is not compliant"). Thus, if a person is on a combination of drugs which is completely suppressing all HIV replication in their body, resistance to the drug cannot develop. However, if such a person were to miss a few doses of their medication, then some HIV replication may occur, allowing the opportunity for the virus to become resistant to the drugs that person is taking. It has been estimated that the virus makes a mistake once every 10 times that it copies itself. Many of these mistakes result in a defective virus which is unable to infect other cells. However, every so often, one of these random mutations results in the virus being resistant to a specific drug.

So, if the person is taking that drug to which HIV has become resistant, then that particular mutant will grow unchecked and become the dominant virus in the person's system. Because there may be up to a billion copies of HIV in the body, many mutations occur everyday if there is high level of HIV present (a high viral load). Thus, one of the major factors causing treatment failure is mutation of HIV, resulting in HIV not suppressed by a particular drug. This most often occurs due to lack of compliance, but also may be due to cross-resistance from prior drug exposure or infection with a resistant strain of HIV, as will be discussed below.

Drug Absorption

Another factor resulting in treatment failure is inability to absorb the drug that a person is taking. In this situation, even though a drug is being taken properly, it is not getting into the blood at adequate levels to completely suppress HIV replication. Due to the many adverse impacts of HIV and opportunistic infections on the gastrointestinal tract, drug absorption can be a major problem for people with HIV infection. If almost no drug is absorbed into the blood, then HIV replication will go on, but drug-resistant HIV will not develop because there needs to be some drug present for a significant drug-resistant population of HIV to develop. However, if some drug is absorbed, then the HIV may become resistant due to suboptimal drug level.

Thus, in the first case, treatment has failed, but not due to HIV drug resistance, while in the latter scenario, HIV drug resistance has resulted from the treatment failure. It is critical to differentiate these two situations. As will be discussed later, resistance to one drug may cause cross-resistance to other drugs in the same class (i.e. PI's). Therefore, in choosing a second drug, it is critical to know whether or not the reason for treatment failure was due to drug resistance.

Drug Activation

Some drugs are administered in a form which requires the body to change the drug into an active form. This may be done, for example if the active form is not absorbed well or is unstable. This is the case for zidovudine (AZT). The body must convert zidovudine into the active form, before it can inhibit the HIV reverse transcriptase enzyme. If the person's own metabolic activation system is not functioning properly (or is genetically different) for the required metabolic conversion, then the drug will not be effective. As with poor drug absorption, a defect in the activation of the drug could result in very low levels or in higher, but still sub-optimal levels. Thus, you might have treatment failure due to very little active drug in the system or due to drug resistance, resulting from sub-optimal levels of the activated form of the drug.

Drug Metabolism

As with drug activation, there are differences in people's own metabolic processes which remove drugs from the body. The rate of absorption and the rate of drug removal determines how much drug is in a person's system and for how long. This is critical for determining the amount and frequency of a drug to take (in order to achieve optimal blood levels) so that HIV suppression is maximized while toxicity is minimized. The most common methods of drug removal from the body are metabolism and/or inactivation in the liver, and excretion by the kidneys into the urine. For some anti-HIV medications, these are very critical processes.

The long list of drug interactions with ritonavir is due to the changes caused by that drug in liver metabolism. The result is that some drugs are metabolized much slower and thus reach much higher levels in the blood, while other drugs are metabolized much faster and thus attain very low blood levels. For example, saquinavir levels are raised by ritonavir. Thus, the usefulness of combining these two drugs is that much higher and more effective blood levels of saquinavir (which is poorly absorbed) can be attained if it is administered with ritonavir.

Conversely, some drugs may be markedly reduced due to the effect of ritonavir on the liver's metabolic processes. These considerations are important to achieve the desired blood level of an anti-HIV drug. Since the effect may vary from individual to individual, ideally, blood levels of the drugs should be measured to determine if, in fact, the drug is getting into the blood at the desired level. This is what is done in early clinical trials (Phase 1 Trials), but it is not routinely done in clinical practice.

Drug-Drug Interactions

Another reason for treatment failure may be due to the direct interaction of two drugs against HIV. Whenever two drugs are used to attack the same bacteria or virus, there may be one of three interactions. The two drugs may work individually, they may augment each other (synergism), or they may inhibit one another (antagonism). So, with synergism, the sum of the two is greater than the individual effect of each added together, whereas with antagonism, the sum of the two is less than the individual effect of each added together. This is why it is important to test drug combinations in clinical trials, because this interaction is not always predictable. A problem today is that because there are so many FDA-approved anti-HIV drugs, it will take many trials and a lot of time to test a new drug in all the possible three, four or five-drug combinations with the currently FDA-approved drugs. Care should be exercised when combining drugs which have not been combined in prior clinical trails.

HIV Drug Resistance

The most common cause of HIV treatment failure is drug resistance, usually due to a combination of the factors discussed above. However, it is also possible that a person may have HIV which is resistant to a drug they have never taken. This may be due to cross-resistance from another drug in the same class (like another PI) that they have taken, or the strain of HIV that the person was infected with was already resistant to the drug, or the HIV developed resistance prior to any drug exposure by random, spontaneous mutation after infection.

Drug resistant usually develops because HIV undergoes a mutation in its genetic material which allows the virus to reproduce in the presence of adequate levels of the drug. In the case of the nucleoside and non-nucleoside reverse transcriptase inhibitors, the genetic mutation causes a change in the structure of the HIV's reverse transcriptase enzyme, which translates the HIV RNA into the host cell's DNA. The result of the mutation is that transcription occurs, even in the presence of the drug, which had previously prevented this process. Similarly, in the case of the protease inhibitors, the genetic mutation results in a change in the HIV-produced protease enzyme which is necessary for assembly of infectious viruses. Testing for drug resistance may be performed by two methods: genotype and phenotype testing.

Phenotype testing is a functional test whereby the HIV is isolated from a person's blood, is placed into a test tube and grown in the laboratory to test its ability to grow in the presence of a particular antiretroviral drug. This test takes several weeks to conduct and is very expensive.

Genotype testing analyzes the genetic sequence of the HIV when isolated from a person's body. Then mutations in the genetic sequence are looked for, which are known to be associated with HIV resistance to a particular antiretroviral drug. This test costs about $300 and can be done quickly.

Currently, neither test is approved by the FDA, and they are not paid for by private or government insurance programs. There is debate among HIV treatment experts about the usefulness of these tests in the clinical treatment of HIV. Currently, many HIV research programs are studying genotype testing.

Cross Resistance

There is now excellent data to show that when a person becomes resistant to saquinavir, ritonavir or indinavir, they are probably resistant to all three of these protease inhibitors. However, some of the data presented at the ICAAC meeting in Toronto in September, 1997 suggested that resistance to nelfinavir may not cause cross-resistance to the other PI's. Other data has suggested that if a person develops a high level of nelfinavir resistance, they will, in fact, also be resistant to the other three PI's. In theory, a person should select a drug in a class (such as PI's) for front line use which has the least possibility of causing cross resistance. Thus, if a person becomes resistant to that first drug, there are some other options available in that class of drugs. Because PI's, nucleoside reverse transcriptase inhibitors and non-nucleoside reverse transcriptase inhibitors all work at a different point to inhibit HIV replication, there is not cross-resistance between the classes of drugs, but rather it is a problem within a class of drugs.

One potentially beneficial use of genotype testing would be when a person has failed a triple-drug combination, testing could provide useful information in determining which of the three drugs the person has become resistant to so that all three drugs would not necessarily have to be abandoned. Mutations identified by genotype testing are reported by the location in the genetic sequence of the virus where the change has occurred (the codon). There are a series of known changes in the genetic sequence of HIV which are associated with resistance in the clinical situation. So, a report might note mutations at codons 41, 67, 70, and 215. These are all sites known to result in resistance to zidovudine.

There are some serious limitations to genotype testing for drug resistance. Drugs such as ddI and d4T do not show genetic mutations which correlate as clearly with clinical resistance as does zidovudine. The genetic change seen which confers resistance to 3TC may actually be beneficial in that the particular genetic change in HIV associated with 3TC resistance confers increased sensitivity to zidovudine (i.e. it makes HIV more likely to be inhibited by zidovudine). With some drugs, only one or two mutations are necessary to make HIV resistant, while other drugs (such as zidovudine and saquinavir) may require several mutations in HIV to confer complete resistance to these drugs (although one mutation at codon 215 will cause complete resistance to zidovudine).

Unfortunately, a report of genotype testing does not give a simple answer to which drugs to use. If that were the case, there would be a stronger call to make the test available to all people living with HIV and for government and private insurers to pay for it. A major limitation of genotype testing is that you need to get a sample of HIV from the patient's blood, and if the viral load is very low this may be difficult. Additionally, there exist in the blood several different populations of HIV, so that some, but not all, of the HIV in your body may be resistant to a drug, but the test may not accurately sample that subpopulation of HIV. The test does not report the number of HIV which contain a specific mutation; it is not a quantitative test. There is a concern that drugs which may be beneficial to a person would be discontinued prematurely (since the presence of a mutation does not mean that the drug is 100% ineffective); or, conversely, drugs which are not helping a person might be continued based on a misleading genotype report showing no mutations present.

Another use for genotype testing for resistance might be prior to initiating any anti-HIV therapy, either in a person recently infected, or someone who has been infected for a long time. One study, reported at the 4th Annual Conference on Retroviruses and Opportunistic Infections, in January, 1997, revealed that in a rural Iowa population, 26% of persons never treated with a protease inhibitor had some genetic mutations associated with resistance to all three of the protease inhibitors which were then approved, and 3% of newly-infected persons had resistance to some reverse transcriptase inhibitors. Other studies have shown that resistance occurs more frequently with lower CD4 counts, and higher viral loads. It appears that the later therapy is begun, the less likely you are to obtain complete viral suppress, allowing for resistance to develop. Thus, the immune system may help in the prevention of drug resistance. This is another observation in support of the , "Hit hard, hit early" approach.

The most comprehensive discussion of HIV drug resistance can be found in the report issued following the International Workshop on HIV Drug Resistance, Treatment Strategies and Eradication, held in St., Petersburg, Florida, June 25-28, 1997. (The summary and other reports are available on the World Wide Web at: http://www.healthcg.com/hiv/hivresistance/content.html.) The main findings were: the realization of generalized resistance across the class of protease inhibitors (cross-resistance), (although as noted above, whether this applies to nelfinavir is not known at this time) highly active antiretroviral therapy (HAART) can suppress viral load levels to very low levels to minimize the chance of resistance developing, even in persons with very low CD4 counts, as long as they were not treated with a lot of antiretrovirals in the past; resistance assays need to be developed to tell us which drugs to use, but at this time, we do not know how to interpret the data; and, antiretroviral therapy needs to be individualized. This report lists, in tabular form, all the known genetic mutations associated with resistance for all the currently available antiretroviral drugs.

One other point emphasized is that once you become resistant to a drug, even if you are off of that drug for one year or longer, the resistant virus is still present and repeat use of that drug will be unsuccessful.

Conclusion

There are many factors which can lead to HIV treatment failure. The most important is the development of resistance to the antiretroviral drug (or drugs). The two most important factors to consider are the amount and nature of prior therapy and compliance with a difficult treatment schedule. An inexpensive, accurate laboratory test which can tell which drug is going to work in an individual person would be very helpful in designing treatment plans. The current genotype and phenotype tests do not yet accomplish this goal. However, for some people, a genotype test may be useful to select antiretroviral drugs, particularly when that person has been treated with many antiretroviral drugs in the past.

980101
STEP9302

Copyright © 1998 - Seattle Treatment Education Project, 1123 East John Street, Seattle, WA 98102. (206) 329-4857 or (877) 597-STEP [7837] (toll-free, valid only in the Pacific Northwest: Washington, Oregon, Idaho, Alaska, and Montana) e-mail: step100@aol.com

ÆGiS is made possible through unrestricted grants from Roxane Laboratories, Inc., iMetrikus, Inc., the National Library of Medicine, and donations from users like you. Always watch for outdated information. This article first appeared in 1998. This material is designed to support, not replace, the relationship that exists between you and your doctor.

ÆGiS presents published material, reprinted with permission and neither endorses nor opposes any material. All information contained on this website, including information relating to health conditions, products, and treatments, is for informational purposes only. It is often presented in summary or aggregate form. It is not meant to be a substitute for the advice provided by your own physician or other medical professionals. Always discuss treatment options with a doctor who specializes in treating HIV.

Copyright ©1985, 2000. ÆGiS & the Sisters of Saint Elizabeth of Hungary. All materials appearing on ÆGiS are protected by copyright as a collective work or compilation under U.S. copyright and other laws and are the property of ÆGIS and the Sisters of Saint. Elizabeth of Hungary, or the party credited as the provider of the content.