(Updated December 2000)
By Tim Horn and Spencer Cox
Introduction
Technology continues to make the fight against HIV a great deal easier. New
tests -- known as HIV drug resistance tests -- are being used by a number of
researchers and healthcare providers and are quickly becoming standardized medical
procedures for people living with the virus.
HIV drug resistance tests are proving to be much like viral load tests when
those first appeared. In 1996 there was a lot of hope -- and confusion -- regarding
viral load technology and how it could be used to help people living with HIV
and to enable their healthcare providers to make better treatment decisions.
Much of the same optimism and confusion can be found today with respect to HIV
drug resistance tests.
This handbook explains how HIV drug resistance tests work and what role they
might play in improving treatment decisions. The question-and-answer (Q&A)
format is designed to answer some of the most frequently asked questions about
the tests. Also included is a
glossary of terms
used in this publication and elsewhere. Finally, a list of additional
resources
is given.
[
back to top ]
What is Drug Resistance?
HIV drug resistance refers to a reduction in the ability of a particular drug
or combination of drugs to block reproduction or "replication" of HIV. For people
infected with the virus, drug resistance can render drugs less effective or
even completely ineffective, thus significantly reducing treatment options.
Resistance typically occurs as a result of changes -- called mutations -- in
HIV's genetic structure (RNA). Mutations of RNA lead to alterations in certain
proteins, most commonly enzymes, that regulate the production of infectious
virus. Mutations are especially common in HIV, as this virus reproduces at an
extraordinary rate and does not contain the proteins needed to correct mistakes
made during copying of the genetic material. HIV relies on many enzymes -- such
as reverse transcriptase, integrase, and protease -- to replicate inside a human
cell. If a mutation of a single site in the reverse transcriptase gene occurs,
the change will remain with the virus as long as it replicates or until another
copying error alters its form yet again. Some mutations cause the virus to become
so weak that it cannot replicate effectively; other mutations may cause the
virus to become even more virulent.
For people infected with the virus, drug resistance can render drugs less
effective or even completely ineffective, thus significantly reducing treatment
options.
Antiretroviral drugs, generally speaking, disrupt the HIV enzyme's ability for
genetic copying, or for making virus that can infect other cells. In a person
who takes antiretroviral drugs, most of the HIV are killed or prevented from
multiplying further. As a result of random mutations that occur on a daily basis,
however, some strains of HIV are naturally resistant to the presence of such
drugs. That is why treatment with monotherapy (a single antiretroviral drug)
is destined to fail.
In essence, drug-resistant mutations are an example of Charles Darwin's principles
of evolution. At first, these particular strains of HIV are fewer than the natural
and most powerful form of HIV -- called the "wild type" -- that dominates the
population. However, once the wild-type virus is destroyed by the offending
drug, the drug-resistant form can reproduce and eventually become the dominant
strain, sometimes within as little as a few days. Thus, only the "fittest" survive,
as in the Darwinian understanding of natural selection.
[
back to top ]
What's With All the Strange Numbers?
A lot of medical information available to both healthcare providers
and people living with HIV frequently discusses specific mutations.
One example is the classic 3TC mutation: M184V. The 184 refers to
the amino acid position on the reverse transcriptase enzyme. The M
-- which stands for methionine -- is the amino acid at position 184
of a wild-type (drug-sensitive) virus' reverse transcriptase enzyme.
The V -- which stands for valine -- refers to the mutation that results
in drug resistance. In other words, the amino acid methionine at position
184 has been replaced by a valine. This change thus prevents an antiretroviral
drug from binding with the enzyme to prevent the virus from replicating.
[
back to top ]
How Can Drug Resistance Be Measured?
Over the past five years, a significant number of breakthroughs have been
made in understanding the power of antiretroviral drugs against HIV.
With the development and availability of viral load tests -- such
as PCR, bDNA, and NASBA -- we can determine from a blood sample how
much virus is replicating in the body. If viral load increases substantially
while a person is on a combination of antiretroviral drugs, the most
likely culprit is drug resistance. Unfortunately, viral load tests
cannot determine whether or not HIV is resistant to one drug in particular
or the entire combination. Moreover, in a person with drug-resistant
HIV, these tests cannot determine which drug or combination of drugs
is likely to be the most effective in the future.
Two general approaches are now used for measuring resistance to HIV drugs.
The first is called genotypic testing. Genotypic tests can help determine
whether specific genetic mutations are causing drug resistance and
drug failure. The second method, called phenotypic resistance testing,
is a more direct measure of resistance and, more specifically, of
the sensitivity of a person's HIV to particular antiretroviral drugs.
[
back to top ]
Can You Explain More About Genotypic Tests?
Genotypic resistance testing examines HIV taken from a patient, looking for the
presence of specific genetic mutations that are known to cause resistance
to certain drugs. For example, it has been well documented by researchers
that 3TC (lamivudine; Epivir) is not effective against strains of HIV
that have a mutation at a particular position -- known as M184V -- in
their reverse transcriptase enzyme (see
What's
With All the Strange Numbers?). If a genotypic resistance test discovers
a mutation at position M184V, chances are that the person's HIV is resistant
to 3TC and is not likely to respond to the drug.
For many drugs -- for example, AZT and protease inhibitors -- complex
patterns of mutations are required for resistance to occur. Interpretation
of these complex patterns can be difficult and incomplete in determining
whether or not the virus is sensitive to particular drugs.
A number of laboratories in the United States and Europe offer genotypic
resistance testing. The most common method of testing uses PCR technology
to make many copies of, or "amplify," the HIV genetic material. Once
amplified, the genetic sequences of particular viral enzymes -- such
as reverse transcriptase and protease--can be examined carefully for
mutations at any of their positions. Depending on the type and number
of mutations found, the researchers may be able to determine whether
someone has developed resistance to a specific drug, since almost
all drugs follow a set pattern of mutations. Some drugs have single
patterns of mutation, but others have complex and unpredictable patterns.
Genotypic
resistance testing examines HIV taken from a patient, looking for
the presence of specific genetic mutations that are known to cause
resistance to certain drugs.
For genotypic tests to be accurate, they generally require the use of
a blood sample from a person who is actively taking antiretroviral medication
and has a viral load higher than 1,000 copies/mL. In the absence of
therapy, the wild-type virus may outgrow the mutant virus. In turn,
the results may not show any drug-resistant mutations, but the mutant
virus may still remain at very low numbers in the person's body and
may quickly increase when therapy with the same drugs is restarted.
[
back to top ]
Are There Disadvantages to Genotypic Resistance Testing?
Genotypic resistance testing has a few advantages over phenotypic testing, most
notably the relative simplicity and speed with which the test can be
performed. The testing can take as little as a few days to complete,
and because it is less complex, it is somewhat cheaper to perform.
Still, some disadvantages of genotypic resistance testing are worth noting.
Most important, it may be difficult to translate the results of a
genotypic resistance test into a meaningful conclusion about the resistance
of the virus to drugs. We have learned a lot about the various genetic
mutations that result in antiretroviral drug resistance, but it is
also true that we don't know
everything about these mutations.
It is possible that we have both over- and underestimated the importance
of specific mutations and their role in causing drug resistance. Moreover,
some genetic mutations have yet to be identified by researchers. Such
is the case with drugs like ddI (didanosine; Videx) and d4T (stavudine;
Zerit). In people who take these two drugs, resistance certainly does
occur. However, researchers are only beginning to determine the exact
genetic mutations that cause HIV to become less sensitive to these
compounds.
Phenotypic testing directly measures the actual sensitivity of HIV to particular
drugs.
This may also be the case with AZT and 3TC. For example, a genotypic
resistance test may demonstrate that a person's HIV has several genetic
mutations that confer resistance to AZT. However, if the person is also
taking 3TC -- which appears to increase the sensitivity of HIV to AZT
-- such genetic mutations may not accurately reflect the amount of AZT
resistance.
There is another disadvantage of genotypic resistance testing. The technology
used to perform the test does not normally evaluate the genetic structure
of small HIV populations found in a blood sample. These small populations
-- called subpopulations -- can contain genetic mutations that do
confer drug resistance. For example, there might be a subpopulation
of HIV that contains a mutation at position M184V (the mutation that
confers resistance to 3TC). Unless this particular strain accounts
for more than 20% of the HIV population found in a blood sample, chances
are that it will not be recognized. Also, genotypic testing may not
recognize strains of virus that are resistant to multiple drugs. For
instance, if a percentage of this 3TC-resistant population also has
genetic mutations that confer complete or partial resistance to AZT
and the protease inhibitor indinavir (Crixivan), genotypic resistance
testing would probably not recognize this small subpopulation in a
blood sample. Should this particular combination of drugs be used,
HIV levels may not be suppressed for long.
[
back to top ]
What About Phenotypic Testing? How Does It Work?
Unlike genotypic testing, which looks for particular genetic mutations that confer
drug resistance, phenotypic testing directly measures the actual sensitivity
of a patient's HIV to particular drugs. To do this, phenotypic tests measure
the concentration of a drug required to inhibit viral replication in the
test tube by a defined amount such as 50% or 95%. This is called IC50
or IC95; IC stands for "inhibitory concentration." Interestingly, this
is the method used by researchers to determine whether a drug might be
effective against HIV before using it in human clinical trials.
Phenotypic resistance testing of HIV is very similar to methods used to measure antibiotic
resistance in bacteria. Sometimes, as is the case of tuberculosis, phenotypic
testing of the bacterium determines whether there are any drugs to which
it will not respond. This procedure has dramatically improved the ability
of healthcare providers to treat such infections effectively. In years
past, the most common way of conducting an HIV phenotypic resistance test
was a slow, labor-intensive process of isolating HIV from a person's peripheral
blood mononuclear cells (PBMC). Many researchers have suggested that phenotypic
testing is a much more direct measure of HIV sensitivity to specific drugs,
but the standard PBMC process is not suitable for routine use and is unavailable
to the majority of patients with HIV. A few companies have developed a
new test in which the key portions of HIV genetic material are amplified
using PCR technology (similar to genotypic testing) so that they can easily
be "inserted" into the shells of laboratory-derived reference strains
of HIV. Sensitivity testing can then be performed relatively quickly under
more standard conditions.
One of the developers of the newer phenotypic technology takes things a step
further by replacing part of the HIV shells with a gene for the enzyme
luciferase. With this enzyme in place, the infected cells glow when the
virus successfully reproduces in the laboratory test. Using light sensors,
the laboratory can then measure the amount of light produced by the virus
in the presence or absence of drugs. Depending on the amount of light
produced -- when compared to that of a wild-type strain of HIV -- the
laboratory can determine both the IC50 and IC95 of the drug.
The concentration of drug necessary to inhibit virus replication is expressed
in units called nanomoles (nM). For example, if the IC50 of the wild-type
virus is 100nM and that of the test virus is 400nM, the test virus is
considered to be fourfold resistant to the drug being tested. In other
words, HIV in the patient is fourfold less sensitive to the drug.
Unlike genotypic tests, the phenotypic resistance test generally does not require
a high viral load. Like genotypic testing, however, it is recommended
that patients be taking antiretroviral therapies at the time of the test.
[
back to top ]
Are There Disadvantages to Phenotypic Resistance Testing?
The results of phenotypic tests are easier to interpret than genotypic tests.
Because phenotypic testing directly measures the sensitivity of the virus
to particular drugs, many researchers and health-care providers have suggested
that these tests are more comprehensive and trustworthy than genotypic
tests. Phenotypic resistance testing procedures are relatively complex,
however, and can take longer than genotypic tests to produce accurate
results -- from ten days to several weeks. The intricacy of these tests
also makes them more expensive.
Like genotypic testing, phenotypic tests are limited in their ability to assess
the drug sensitivity of HIV subpopulations. Again, this may prevent such
tests from producing an accurate estimate of HIV's ability to respond
-- especially for long periods of time -- to specific antiretroviral drugs.
Another challenge is that we still do not fully understand what level of resistance
translates into a failure of treatment. For example,
a five-,
six-, or sevenfold reduction in the sensitivity of HIV to
a protease inhibitor is considered "moderate." But is there a significant
difference between a fivefold reduction and a sevenfold reduction? We
need to learn what level of resistance tells us that a given drug is no
longer useful.
[
back to top ]
Might the Tests Work Together to Overcome Their Individual Disadvantages?
Perhaps. As explained above, each type of test has its limitations. Knowing which
genetic mutations are present may not accurately predict HIV's sensitivity
to a particular drug; however, phenotypic tests are slower and more
expensive. Using both tests together could certainly help deal with
some of the weaknesses of each test administered individually.
Using both genotypic and phenotypic tests together could certainly help deal
with some of the weaknesses of each test administered individually.
Both tests are slightly limited in their ability to measure subpopulations
of potentially drug-resistant HIV. Using both tests together may not
necessarily overcome this obstacle. Moreover, it is important to recognize
that solid evidence proving the utility of these tests needs to be generated
for a number of clinical situations. Which treatments should be switched
to in the event of failure? Further clinical trials and research will
attempt to answer these questions.
[
back to top ]
How Can These Tests Help Decide on an Initial Treatment Regimen?
Based on what is known about HIV's error-prone replication process (see above),
we can assume that all patients have at least a few subpopulations of
HIV that are resistant to individual drugs. However, these strains are
often too limited in number and strength to compete with wild-type virus,
and they stand a good chance of being killed off by initiating combination
antiretroviral therapy. After all, the purpose of combination therapy
is to serve as a multipronged attack on such strains.
People harboring multiple-drug-resistant virus can transmit it to others.
A potential threat, however, is the
transmission of multiple-drug-resistant
strains of HIV. Multiple-drug-resistant HIV (MDR-HIV) is defined as
a strain of the virus that has limited or no sensitivity to several
antiretroviral drugs. Such viruses usually emerge in HIV-infected people
who were not prescribed drugs in the optimal way or who were not able
to adhere to the challenging demands of drug-taking schedules. People
harboring such virus can then transmit it to others.
Some researchers have found that HIV is either partially or fully resistant
to one or more of the commonly used antiretrovirals in between 10% and
30% of newly infected people. Such cases are likely to increase dramatically
in the near future.
For instance, in a recent study from San Diego, California, published in
the
Journal of the American Medical Association (
JAMA),
141 patients who had become infected with HIV in the previous year and
had received less than seven days of anti-HIV treatment were tested
for drug resistance. Some resistance to at least one anti-HIV drug was
found in 36 patients, or more than 25% of the study participants. Two
percent of patients had substantial resistance to at least one drug.
In another study conducted in New York and also published in
JAMA,
80 newly HIV-infected people were tested for drug resistance. About
27% of the patients had some evidence of drug resistance, and resistance
to several drugs was found in almost 4% of participants.
Although presently neither genotypic nor phenotypic testing is being used by
many healthcare providers for this purpose, recently released federal
guidelines state that use of resistance testing be considered for selecting
an initial treatment regimen.
[
back to top ]
How Can These Tests Help in Choosing a New Treatment Regimen When an Old One Fails?
Drug failure is loosely defined as an increase in viral load, a decrease
in T cell counts, and/or signs of physical disease progression in people
who are on combination antiretroviral therapy. Although drug failure
can also be used to describe the experience of people who must stop
their medication because of intolerable side effects, it is most often
associated with the presence of genetic mutations and decreased drug
sensitivity.
Viral load tests are likely to remain the most important tool for determining
whether or not drug failure is occurring. Drug resistance tests, on
the other hand, may play an invaluable role in helping doctors and their
patients understand why failure has occurred and what treatment options
are still available.
[
back to top ]
Are There Any Results From Clinical Studies?
A number of clinical studies of drug resistance testing are currently
being conducted in the United States and abroad. The results of some
studies have already been presented, both at major medical conferences
and in medical journals.
To date, several retrospective and prospective studies have been completed.
In a prospective study, volunteers are enrolled specifically to assess
the effects of a particular treatment, medical procedure, or test. A
retrospective study is an analysis of data from a study that has already
been completed. And while several retrospective studies have demonstrated
that drug resistance tests were very useful, the best way to determine
if these assays will work in the real world is by conducting prospective
studies.
In mid-1999, results from the first prospective study involving genotypic
resistance testing were published in the journal
Lancet. In the
study, which was conducted in France, HIV-positive people who were seeing
their viral load increase while on a triple-drug regimen were randomly
assigned to one of two groups. Only those in group B (65 people in total)
were permitted to make treatment changes based on the results of resistance
testing; patients in group A (43 people in total) determined which salvage
therapy they should take based on viral load only.
After three months, the people in group B had viral loads that were significantly
lower than those in group A (-1.04 vs. -0.46 logs, respectively). After
six months of therapy, people in group B were almost twice as likely
to have undetectable viral loads (32% vs. 14%, respectively). According
to the team of researchers conducting the study, these results suggest
that drug resistance testing was of significant help in determining
which drugs should be used in a salvage regimen to yield more effective
results. The team also acknowledges that the results are from a relatively
small, short-term study and that much more information is needed from
additional clinical trials.
Some researchers have found that HIV is either partially or fully resistant
to one or more of the commonly used antiretroviral in between 10% and
30% of newly infected people. Such cases are likely to increase dramatically
in the near future.
With respect to phenotypic testing, the results from the first prospective
study were recently presented at a major medical conference -- the 7th
Conference on Retroviruses and Opportunistic Infections -- held in early
February 2000 in San Francisco. Similar to the prospective genotypic
testing study discussed above, HIV-positive people who were failing
an anti-HIV drug combination were randomized to one of two groups: one
group (144 patients in total) was permitted to pick a new combination
of drugs based on the results of phenotypic resistance testing, whereas
volunteers in the second group (130 patients in total) determined which
treatments they should switch to based on the anti-HIV drugs they had
tried in the past.
Using a conservative type of data analysis -- called the intent-to-treat analysis
-- patients who were permitted to use phenotypic testing appeared to
do better than those who simply chose new drugs based on their treatment
history. Overall, 38% of patients who used phenotypic testing to plan
a new drug regimen had undetectable viral loads (< 400 copies/mL)
after 16 weeks on their new combination. Among patients not permitted
to use phenotypic resistance testing, only 22% had undetectable viral
loads after four months.
[
back to top ]
Would It Be Necessary to Switch Every Drug?
At the present time, it is recommended that anyone who appears to be failing
a particular combination should switch to an entirely new batch of drugs.
This, of course, is frustrating, as many HIV-positive people do not
have three or more untried drugs from which to choose. It may also be
a wasteful decision for those who do have several remaining options.
Why toss out a drug that may, in fact, still be extremely effective
against HIV?
With drug resistance testing, it might be possible to weed out the ineffective
drug or drugs in a given combination.
With drug resistance testing, it might be possible to weed out the ineffective
drug or drugs in a given combination. For example, in a study published
in
JAMA in January 2000 involving people taking an antiretroviral
combination of indinavir, AZT, and 3TC, 17 patients experienced viral
load increases while receiving therapy. Although it would make sense
to blame such viral load increases on multiple-drug resistance, resistance
tests demonstrated that 14 patients had developed resistance to 3TC
only; HIV in these patients could generally still be blocked by indinavir.
[
back to top ]
Do Most Experts Recommend These Tests?
Yes. Two important groups of medical experts now recommend that drug resistance
tests be used in helping HIV-positive people plan their treatment regimens,
especially if a switch in therapies is needed. One group that recommends
drug resistance testing is the United States Department of Health and
Human Services (DHHS), a branch of the federal government that oversees
public health in the United States. A second group that recommends these
tests is the International AIDS Society-USA (IAS-USA), a private medical
organization made up of many leading HIV/AIDS experts in the United
States and elsewhere.
[
back to top ]
Where Can These Tests Be Accessed?
Resistance tests are generally available through various clinical reference laboratories
and hospitals. Genotypic tests typically cost between $300 and $500,
whereas the phenotypic tests cost between $700 and $900.
[
back to top ]
Will My Insurance Company, ADAP, or Medicaid Cover Their Cost?
Although these tests are becoming widely available, insurance coverage is highly
variable and may depend on where you live and what health coverage
is offered. Local AIDS service organizations should be able to provide
information about the willingness of government-funded programs to
pay for resistance tests.
Now that drug resistance tests have been recommended by the DHHS and IAS-USA,
it is likely that more insurance companies and other third-party reimbursement
programs will pay for the tests.
[
back to top ]
About the Authors
Tim Horn is Executive Editor of the
PRN Notebook, published by the Physicians'
Research Network in New York City. He also writes for Medscape.com
and AIDSmeds.com, and has worked for a number of other AIDS organizations,
including the American Foundation for AIDS Research (amFAR), the AIDS
Treatment Data Network, and the PWA Health Group.
Spencer Cox is a longtime advocate for people with HIV/AIDS.
Douglas Richman, M.D., is Director of the Research Center for AIDS and
HIV Infection at the San Diego Veterans Affairs Medical Center and
Co-Director of the University of California San Diego Center for AIDS
Research.
Important note: Information in this article was accurate in 2001. The state of the art may have changed since the publication date.
This information is designed to support, not replace, the relationship that exists between you and your doctor.
