Phenotypic Assays Become Widely Available
The article by Sarah Palmer in this edition of Treatment Issues refers to the use of cell culture assays to determine what drugs a particular HIV isolate can resist. By comparing the isolate's activity in the presence or absence of a series of drugs, researchers can construct an accurate resistance profile for that isolate. Such "phenotype" analysis has long been considered the gold standard for determining resistance since it displays exactly what that resistance is.
Phenotypic tests typically involved up to six weeks of cell culture and were very expensive. They therefore have not been available for use by individual patients and their doctors. Instead, doctors and patients have relied to a certain extent on the cheaper and quicker genotypic testing that took a patient's HIV and tested for the presence of certain mutations.
Since our knowledge concerning which mutations give rise to what kind of drug resistance remains very incomplete, genotypic testing has obvious drawbacks. Critical mutations are missed because their importance has not yet been established. The importance of other mutations might be over- or under-estimated because of misunderstandings about their individual role or the way they interact with other mutations. Resistance to such drugs as ddI and d4T is poorly mapped out and frequently appears without mutations known to be relevant to those particular drugs. (Again, see Sarah Palmer's article concerning the way broadly cross-resistant HIV can arise after exposure to a subset of the available drugs.)
Finally, genotypic testing misses the gene makeup of minor HIV subpopulations in a person's body. These subpopulations can contain mutations that give them an advantage during treatment with a particular antiviral medication. When the wrong drug is administered to a patient, they will continue to replicate, becoming the dominant HIV strain in the body and render the therapy ineffective. All these criticisms still hold for more advanced genotype assays, such as Affymetrix's "gene chip," which sequence entire genes and genomes. Here, one may have a read out of all the differences between a patient's HIV and a consensus "wild type" gene set, but interpretation of the significance of those differences remains difficult and minor subpopulations may still be invisible.
Now, two adaptations of phenotypic testing are becoming widely available. Representing improvements in the ease and cost of classical phenotypic assays, they overcome many, but not all, of the inaccuracies of current genotypic assays.
The Virco Antivirogram
This July, LabCorp, one of the major US medical laboratory chains, will start offering a phenotype test developed by the Belgian company Virco. The Virco test is known by the trademarked name "Antivirogram." It originated about five years ago in the British laboratories of Burroughs Wellcome, before that company was taken over by Glaxo. The Antivirogram has been available for the past two years through a handful of American doctors who had established their own independent relation with Virco. Because of the ad hoc connection with the Belgian site, adequate handling of the patient's blood sampling was not guaranteed, shipping the samples abroad added considerable extra expense, and four-week delays in receiving test results were frequent.
LabCorp will perform some of the initial, preparatory steps on patients' blood specimens and then send them in a group to Belgium for most of the processing. The assay, which costs $800 to $1,000, takes about three weeks to complete. Virco hopes to reduce this interval to about 12 days in the near future.
The present Antivirogram covers all marketed reverse transcriptase and protease inhibitors plus abacavir. More drugs of the same classes could be added at any time, although this increases the test's expense. The test itself is very complicated and has to be done in a special laboratory. It corrects for some of the nebulousness of conventional culture assays by employing a standard HIV virus as the foundation for gauging HIV growth kinetics.
This virus's reverse transcriptase and protease genes are removed and replaced by the equivalent genes separated from the patient's HIV and multiplied using PCR techniques (similar to what is done in the Roche viral load test). The reference strain serving as the carrier virus was selected for its consistently rapid, but still controllable, growth properties. (A major headache with traditional culture tests was that viral isolates would grow either too slowly or so fast that differences in drug sensitivity could not be detected.)
After a week to ten days of culturing the hybrid HIV, the virus is harvested and fed to an automated process that grows the virus in the presence of different antiviral drugs at various levels and checks the resulting viral replication.
The phenotype results are reported in chart and table form comparing the IC50 determined for each drug with the recombined patient-reference HIV to that for the original reference strain alone. (The IC50 is the drug concentration needed to reduce HIV activity in cell culture by half. The higher it is, the more resistance HIV is exhibiting to that drug.) This comparison yields the "fold resistance" for each drug (how many times that drug's IC50 has been increased in the hybrid HIV relative to the reference strain).
Displayed too is whether the results indicate continued sensitivity to each drug (less than four-fold increase in the IC50), intermediate resistance (four- to ten-fold) or a high degree of resistance (over ten-fold).
A Rapidly Moving Newcomer with Another Fancy Name
A year ago, the ViroLogic corporation only had a patent for a new phenotypic resistance assay. Today, the company has a building in South San Francisco complete with offices, labs and two robotic testing lines in the basement. In the coming months, ViroLogic will begin to commercialize its assay to researchers and clinicians, although capacity will be limited at first.
ViroLogic's phenotype test works in similar fashion to the Virco one -- producing hybrid viruses based on a standard reference virus combined with the reverse transcriptase and protease genes from a patient's HIV. The difference is that this recombined HIV "test vector" is then further hybridized by replacing part of its envelope gene with the gene for firefly phosphorescence (the luciferase enzyme) and surrounded with a functional envelope from murine leukemia virus. Since the HIV envelope gene in the resulting virus is defective, this virus can only infect new cells and reproduce for one generation in the test tube. When replication is successful, the new HIV virions produce quite a sparkle, though, and this luminescence can be measured by light sensors.
The reduction in luminescence produced in the presence of different doses of each antiviral drug is compared to the reference HIV's response to the same doses of each drug. As with the Virco test, comparing the standard HIV and patient HIV IC50s for each drug indicates the degree of resistance present in the patient's HIV. Since only one viral generation is involved in the test's final phase, the assay proceeds more quickly than is true for the Virco test at present. ViroLogic's turnaround time is seven to ten days at present.
Do We Need an Extra Test?
In the traditional spirit of the marketplace, ViroLogic's price is expected to be a little under what Virco charges. Considering all the other testing already conducted on or proposed for people with HIV, one might well wonder if phenotypic resistance assays are worth the added expense and effort.
The tests remain hobbled by the fact that although they do catch mixed HIV populations, they still miss minor viral strains comprising less than 10% to 20% of a person's total HIV. They also require that someone have a significant viral load so that there is enough HIV to extract and test. Lastly, they do not check the performance of combination therapies. Detecting "intermediate" resistance to a certain drug may not rule it out as part of an individual's combination regimen, but only if no HIV isolates highly resistant to that drug are hiding in the population mix.
For these reasons, these assays cannot be used to predict the imminent emergence of drug resistance. Their immediate use will be as an aid in understanding the cause of antiviral treatment failure in an individual. They have had mixed results when using them to choose salvage therapies. Here are two examples:
There is nothing that forces a focused study of these tests' role in treatment management: The FDA does not regulate laboratory tests. The agency only intervenes in the sale of equipment kits that enable local medical labs to perform specific tests (as is the case for viral load testing). Virco's and ViroLogic's tests will always be confined to the companies' own specialized laboratories because of their technical complexity and the safety issues raised when culturing HIV.
All is not completely lost, however. Up to now, intrepid patients have had to pay for resistance testing out of their own pocket. To market their tests, Virco and ViroLogic will both have to campaign to obtain coverage by insurance companies, HMOs and Medicaid. They will need to demonstrate that the testing, while expensive, saves money and improves patient outcome as compared to just using viral load tests to judge patient's response to therapy. As the two companies gear up for this effort, they will have to generate data that are at least persuasive, if not as rigorous as would be required for a formal FDA review.
To this end, both Virco and ViroLogic are building up databases relating findings from their tests to later treatment outcomes. Major validation studies are still in the discussion stage, though. Such studies could be retrospective, using stored samples from previous clinical trials whose outcomes are already documented, or prospective ones, in which new study participants are recruited and the benefit of resistance testing for patient management is recorded as it unfolds.
This article was provided by Gay Men's Health Crisis. It is a part of the publication GMHC Treatment Issues. Visit GMHC's website to find out more about their activities, publications and services.