Robert Shafer What is the relative benefit of genotypic versus phenotypic tests? Should both be obtained? If only one is allowed by a third party payer, which should be used?
Mark Wainberg:
Right now, the phenotypic assays are so expensive, and there just aren't that many places around that are doing them in the kind of reproducible way that the companies want, that we have to say for now that the genotypic assays are not only going to be on board first. Also their relatively low cost makes them a priority, I'm sure, for third party payers. Since we want to encourage the development of the field (and I think we have to walk before we run), and given as well, I think, the fact that we now have two studies (GART and VIRADAPT) showing, at least in restricted settings, the value of doing genotyping, coupled with expert consultation perhaps, I think we're inching forward to the era when we'll be recommending genotypes, assuming that third party payers are going to be willing to go for them. Expert outside opinion will be needed for interpretation of results in the context of a particular individual and/or treatment history.
David Katzenstein:
Mike, you have experience with both tests. You were saying your clinicians in some ways prefer the ease of interpretation of phenotype. But I'm wondering if, looking at what's going on in the clinic now, you have a sense of which might work best.
Robert Shafer:
My observation, I think, is that everybody is complaining about the resistance tests being so difficult to interpret, and that's true. I mean, it is very difficult to interpret. It's very hard to make any sense of it. But a lot of physicians have been ordering these for two years, and in some way have become much more comfortable with the much more complicated test, just because they've been using it for two years.
Mark Wainberg:
By "more complicated" do you mean genotypic?
Robert Shafer:
The genotypic, yes. But there's a certain level of familiarity that's developed by physicians using a test, and I think that's something forgotten, that in the practice of medicine, a lot of what people are doing, whether it's right or wrong, whether it's evidence-based medicine or not, there's a certain comfort level in continuing what you're doing. And now that the phenotypic tests are coming along, you'd think that physicians would be jumping to a test that gives an easier result to understand, and that there's a persuasive argument why a phenotypic test should be used. But I find that unless the two are brought along together and are able to be compared side to side, it's going to be very hard for physicians and clinicians to jump from genotype to phenotype based on any particular argument. I think they're going to have to coexist for a while. I think there are going to have to be head-to-head studies, and that's something which, as far as I know, has been resisted by certain of the players in the field. And so, essentially, I'm agreeing with Mark that we're early in the process and more studies need to be done, and genotype has a head start. Phenotypic tests, I think, have to be used in conjunction with genotype tests for the next few years, to figure out the role of each.
Mark Wainberg:
Let me just make one more point, to touch on something you said in regard to another interesting subject. It doesn't touch on this issue as directly as on another interesting subject. Notably, it would be awfully interesting to have a profile of the physicians who have been requesting genotypes during the last couple of years. I suspect that they would tend to be relatively expert (shall we say) in regard to what resistance testing means, and perhaps as well in regard to how to interpret genotypic data once they're available. So we might not have a fairly accurate cross section of the population of people demanding these tests or requesting these tests at this time.
Brian Conway:
[Post hoc addition.] In situations where the number of mutations is small and the genetic background contains few polymorphisms, it is likely that the phenotype can be inferred from the genotype, and only the genotypic test is needed. As soon as this is not the case, it would be best to perform the phenotypic assay to best guide therapeutic decision-making. This is particularly true is salvage therapy, but may be less true in the evaluation of untreated patients or those having a virologic breakthrough after the first therapy, where the drug pressure that has been applied is more "straightforward".
A good scenario for third party payers is that a genotype would be authorized in all cases, but an expert review is mandated before a phenotype is ordered and done. This is the responsibility of the ordering physician, the same as when an MRI is ordered (at least in Canada!!!) instead of a CT, for example. The sample would have to be collected with this in mind, so that the sample is readily available when the phenotypic testing is approved.
