HIV Drug Resistance Testing and Update
from Resistance Workshop- San Diego June 23
is Genotypic Resistance Testing?
There are 2 types of resistance testing-genotypic testing and
Genotypic resistance testing looks at a person's HIV genes (enzymes) to detect if there have been changes to it. These changes are called mutations.
Generally, changes in these HIV genes are called mutations, and when these changes occur HIV drugs are not as effective. There are certain mutations that
are known to be associated with resistance to specific drugs. Mutations to the protease enzyme can result in a protease inhibitor not binding to the protease
enzyme as well. The protease inhibitor has to bind to the enzyme to prevent the protease enzyme from doing its job in the production process of HIV.
Resistance leads to drug failure (viral load rebound). Less is known about how resistance to AZT develops. The emergence of certain known mutations leads
to AZT resistance. AZT and other NRTIs (nucleoside reverse transcriptase inhibitors) inhibit the reverse transcriptase enzyme from performing its role in viral
replication. When AZT mutations develop, AZT is less able to inhibit the reverse transcriptase enzyme, and permits viral replication. For example, if your HIV
viral gene changes (mutations) at positions 70, 215, and 219 researchers know this means you may have resistance to AZT. But the process appears to be
more complicated and is not well understood. Although a person can fail d4T and ddI, mutations are not necessarily detectable. Resistance to 3TC is clear
when the 184 mutation is seen. Regarding protease inhibitors, if you have mutations in the protease gene at positions 48 and 90 you have resistance to
saquinavir. For indinavir any combination of several mutations from a group of a number of well known mutations indicate resistance to indinavir. So genotypic
resistance looks to see if you have mutations known to be associated with resistance to specific drugs. In effect it is an indirect measure of resistance.
Resistance to efavirenz (Sustiva) is associated with the K103N mutation. If your genotypic resistance test reports a K103N mutation, you have resistance to
efavirenz. Generally, nevirapine resistance is also associated with the 103 mutation. As wel, additional NNRTI mutations can develop. Generally, the
development of the 103 mutation results in cross-resistance to other available NNRTIs. Improper adherence, such as missing doses or eating fatty foods with
indinavir, is a leading cause of resistance. People can fail a drug for other reasons than resistance. For inexplicable reasons, a person may not reach and
maintain adequate blood levels of a drug. Their metabolism could theoretically clear a drug too quickly.
(You want to use a reliable lab)
Virco - based in Europe but available in the USA only
through LabCorp. Virco provides their own phenotypic test (Antivirogram)and they use the
ABI test for genotyping (VircoGEN). Samples are shipped to LabCorp and then to Virco for
ViroLogic - based in South San Francisco, CA. They
provide a phenotypic test (PhotoSense)now. Plan to have genotypic test in future.
Visible Genetics - provides only a genotypic test.
Located in Norcross, GA.
Resistance testing can be used to check for pre-existing
resistance prior to starting any therapy. This would be to check for resistant virus
transmitted by sex or IV drug use.
What is Phenotypic Testing?
With Phenotypic testing-you take virus from a person and put
it in a test tube and add increasing amounts of a drug. You want to see how much the virus
is able to reproduce itself as you add increasing amounts of drug. And how much drug is
required to stop virus reproduction. If all you need to stop HIV from reproducing is an
amount of drug equivalent to the amount a person would take that person does not have
phenotypic resistance and that drug should work for that person. If you need 4 times that
amount of drug to suppress HIV from reproducing then your virus is 4-fold resistant to
that drug. If you need 10 times the amount of drug than your virus has 10 fold resistance
to that drug. Exception ABT-378 because drug levels are very high.
Phenotypic testing actually measures more directly if you are
resistant to a drug. Several studies have been reported, a few at the Resistance Mtg, that
if you detect genotypic resistance that may be predictive of phenotypic resistance but
possibly not always. Virus with 5-fold phenotypic resistance had no genotypic resistance
in 4 cases (n=38).
& Weaknesses of Genotypic & Phenotypic Testing
- Genotypic testing reports to you the actual mutations
detected. For example, if you have indinavir resistance it should detect any of these--82,
46, 84, 63, 90 and possibly additional ones. The major mutation causing resistance to
nelfinavir is D30N. Ritonavir primary mutations include 82, 84, 54. Fortovase major
mutations include 48 & 90. You will see primary mutations and secondary mutations.
Primary mutations are enough to cause resistance.
- Some doctors prefer phenotypic testing because the presence of
genotypic mutations does not necessarily mean you have full resistance to a
drug.Phenotyping will detect actual resistance to a drug within limitations rather than
seeing just a list of mutations.
- But several studies presented at Resistance Workshop showed
and suggest that genotypic resistance can predict phenotypic resistance but you have to
know how to read the results and the lab has to know how to interpret the results in their
report to the doctor.
- Genotypic testing takes a shorter time to get results back. I
hear it can take a week to 10 days. In the future it may take a few days. For phenotypic
testing Virologics says 2 weeks; Virco says 4 weeks.
- This applies to both genotypic & phenotypic testing--if
too few of your viruses, so-called minority species, have resistance (<10-20%) the
resistance tests may not pick it up. This means resistance or mutations may be below the
surface or ability to detect them but they can emerge or evolve over time, sometimes
- Bor both geno- and pheno- its recommended that the best time
to perform tests is while on drugs. If you perform tests a month or longer after stopping
drug you may not be able to detect mutation or resistance. But that doesn't mean its not
below the surface, just it may not be detectable--as described above.
- In a few studies people who were on for example amprenavir or
indinavir+AZT/3TC, had a viral load rebound but only the 3TC resistance mutation was
detected--M184V. No protease inhibitor mutations were detected. Some researchers suggested
you should stay on your PI and switch the other drugs. My opinion is I think that's too
risky. Richard D'Aquila, a noted resistance researcher, is switching the entire regimen
and I think that's the safer.
- In general, resistance testing may be useful to detect
non-adherence. If a person's viral load is detectable but there is no detectable
resistance the person may not be taking their drugs and the doctor has the opportunity to
ask the patient. You should be honest with your doctor about this.
- Some mutational interactions are not well known and some
haven't been identified yet. For example, adefovir is more potent if the 3TC M184V
mutation is present. Some doctors may not know this. M184V may reverse AZT resistance for
some people for a period of time.-- Some of these types of relationships may exist but
have not yet been identified
- Regarding genotypic tests--The test results may not be
interpreted properly by the lab performing the test. They may send back to the doctor a
misinterpretation of the meaning of the mutations. The doctor may not be able to
understand the report because he/she may not understand the meaning of specific mutations
or because it's unclear how prevalent a mutation may be.
- A genotypic mutation can occur that has never been seen
before, and so it is not known whether it may or may not contribute to resistance for any
given drug. Phenotypic resistance, however, detects this but this is a limitation to
genotypic testing. For example, the 69 mutation was only discovered last year.
- A study was conducted where the same samples were sent to
various labs but they did not receive back the same results from all labs. In fact the
variations were worrisome. So its important to select a lab in which you have confidence.
You may receive different results from different labs. This survey also pointed out some
of the difficulties in interpreting genotypic test results. Phenotypic testing can, in my
opinion, address some of these concerns. But phenotypic testing can take longer to
perform, they are more expensive, and there is preliminary data suggesting that genotypic
resistance may predict phenotypic resistance. And 2 studies using only genotypic testing,
Viradapt and GART, have shown that better treatment decisions were generally made using
genotypic testing. This was measured by changes in viral load. People in these studies who
used genotypic testing in making treatment decisions had greater reductions in viral load
than study participants who did not use resistance testing in making the treatment
decisions. In the GART study treatment decisions were made by two means: genotypic testing
and expert advice. The expert advice component is important because an expertise is very
important in interpreting using genotyoic test results in optimizing treatments decisions.
- The amount of resistance required for failure of a specific
drug may not be known yet for all drugs. This applies to ABT-378. For example it is
generally accepted that 4-fold phenotypic resistance is equal to drug resistance and the
phenotypic reports you get back say you have <4-fold, 4-10 fold, or >10 fold
resistance. Because ABT-378 achieves very high drug levels, which is a reason for its
potency, it may take a higher level of resistance to ABT-378 to see actual drug failure.
- What does 4-fold phenotypic resistance actually mean?? The
IC50 stands for the inhibitory concentration of drug necessary to stop 50% of drug
replication. A certain amount of drug is required to inhibit 50% of drug replication. If
there is no drug resistance the amount of drug required to inhibit 50% of replication is
less than if resistance is present. If you need four times as much drug to suppress 50% of
replication than you have four-fold phenotypic resistance. First they test how much drug
is necessary to accomplish this against a virus with no resistance, a so-called wild-type
virus. This serves as a reference for measuring how much drug it takes to suppress 50% of
replication in a sample submitted by a person to tesat their resistance.
- COST-- a phenotypic test can cost $800 while a genotypic test
can cost $400 or less, if insurance doesn't cover it.
- The tests are better at detecting resistance to a drug than
sensitivity to a drug. That's because if you detect mutations you know they are there but
if you don't see a mutation it may still be there below the surface.
Showing Benefits of Resistance Testing
There have been a number of studies. Following are two
high-profile studies I selected.The follow-up results of these 2 were just reported at
- VIRADAPT ..study-- 108
people failing therapy (VL >10,000 copies/ml) of NRTIs (>6 mos) + protease inhibitor
(>3 mos) were randomized to standard of care or treatment according to genotypic
resistance mutations (Visible Genetics). People had used an average of 4 NRTIs (39 mos)
and 1.8 PIs (11 mos). After 6 months, the people using genotyping had a 1.15 log reduction
in viral load vs a 0.67 log reduction in other group. 32% vs 14% had <200 copies/ml.
People in standard of care group then were offered genotyping to make treatment decisions.
At months 9 & 12, in the original genotyping arm the viral load reductions were
maintained at about 1.15 log, but now the 30/43 (69%) who accepted the use of genotyping
their viral load reduction was about 0.87 log at 9 & 12 mos. The percent below 200
copies/ml rose from 14% at month 6 to 25.7% at month 1
(genotypic anti-retroviral resistance testing)
- GART .study- 153 people
who had detectable viral load (28,000 copies/ml) were randomized to using genotypic
testing and expert advise or made standard of care treatment decisions. People had good
amount of previous drug use- 82 failed indinavir, 51 nelfinavir, 11 ritonavir, 9
saquinavir; 73% had at least major NRTI mutation and at least 1 primary PI mutation.They
looked at the change in viral load at average of 4 & 8 weeks: GART arm---1.19 log vs
no-GART -0.61 log VL reduction. The difference between the 2 groups at week 12 was -0.44
of Transmitted Resistant Virus
One study of 133 people (56 days after seroconversion within
last 12 months in USA) reported major resistance among 3% (using ViroLogic phenotype
testing- >10-fold-- and ABI genotype testing):
- 2% to NRTIs
- 1% to NNRTIs
- 2% to protease inhibitors
- mutations observed (215, 184, 90, 46, 82, 84) plus secondary
Moderate reduction in susceptibility or moderate resistance
(2.5 to 10-fold resistance) to 1 or more drugs among 29%--
- 4% NRTIs
- 18% NNRTIs*
- 11% protease inhibitors
* (numerous polymorphisms but no primary drug mutations)
A second study reported at Resistance Workshop on 230 samples
using VircoGEN and Antivirogram.
- Primary mutations noted -215, 103, 184, 190, 48, 82, 84 +
For genotype, resistant (R) (>10 fold) and intermediate
(I) (4-10 fold)-
- 2% NRTIs (R) - 5% (I)
- 6% NNRTIs (R) - 42.2% (I)*
- 2% protease (R) - 38.2% (I)
For phenotype testing-
- 1.6% NRTIs (R) - 4.7% (I)
- 7.3% NNRTIs (R) - 21.9% (I)*
- 1.6% Protease (R) - 3.1% (I)
1% were resistant to 2 classes of drugs. 10% had significant
phenotypic change or ant genotypic change.