Pre-Conference Genotypic Resistance Test Seminar
From Jules Levin
So far, my visit to Glasgow has been thoroughly enjoyable, unlike, Geneva and other cities I visited for AIDS conferences. The people, food and service in Glasgow have been great. But Scottish prepared foods tend to be rich, which is not good for my lipids levels. Visible Genetics, Inc sponsored a symposium on Saturday evening, Nov 7 prior to the start of the official opening of the Conference on Sunday evening. They took this opportunity to report a preview of the interim 6 months data from a prospective study called VIRADADT comparing viral load changes of those who used the Visible Genetics genotyping test to select a salvage regimen to those who did not. This data wil be the subject of an oral presentation at the Conference. The data set available now includes 47 of the 108 patients participating in this randomized, controlled clinical trial being conducted by Dr Pierre Dellamonica at Centre Hospitalier Universitaire de Nice.
The underlying premise is that drugs for which a person does not have resistance may be selected for a person's regimen by genotyping HIV. Several studies presented at Geneva showed retrospective data suggesting that genotyping and phenotyping may be able to predict which drugs should or should not be selected to use in a salvage regimen. All the patients entering VIRADAPT averaged 4.7 (50,000 copies/ml) baseline viral load, had failed at least one triple therapy regimen, and were taking at least one protease inhibitor. After 6 months it was reported that 39.1% of the paients using genotyping to select their regimen had <400 copies/ml while 9.5% of the individuals who didn't use genotyping (control group) had <400 copies/ml. The viral load reduction was 0.40 log in the control group and 1.20 log in the group using genotyping. Of those individuals who used ritonavir+saquinavir in their salvage regimen, individuals using the genotyping had a 1.20 log reduction, and those in the control group had a 0.60 log reduction. It was reported that 41.9% in arm A had a firstline regimen failure, and 58.5% in arm B had a first line regimen failure. While 37% in arm A had a second line regimen failure and 20% in arm B had a second line regimen failure. These data are preliminary as an additional 61/108 of the study participants have not yet reached the 6 month analysis point. Further testing is required to confirm these results.
Dr Charles Boucher is a resistance researcher at Utrecht University in the Netherlands. He discussed his research on evaluating the accuracy and sensitivity of genotyping done by 23 labs (11 in Europe and 12 in the USA). Based on this research, he said currently there is inadequate quality controls in general in laboratories conducting genotyping. He sent a panel of "well-defined" blinded samples containing drug resistance mutations at codons 41, 215 and 184 at different concentrations relative to wild-type to 23 labs worldwide. The detection of the mutant genotypes was determined qualitatively and quantitatively.
Boucher reported that all labs detected the presence of sequence heterogeneities at 41, 215 and 184 in one or more of the panel samples, though not all reported the correct codon genotype. Two labs reported a mutant genotype in samples containing only the wild type, while 2 and 3 labs respectively, failed to detect the mutant 41 and 215 genotype in a complete mutant DNA population. Mutations present at relative concentrations of 25% of the total DNA population were successfully identified by only 13/23, 10/23 and 16/23 labs for codons 41, 215 and 184Val. Regarding the 215 mutation in the latter example where 10/23 successfully identified the 215 mutation, 11/23 labs reported no mutant detected and 2/23 reported an incorrect codon. Regarding the 41 mutation, 8/23 did not detect a mutation, and 1/23 reported an incorrect codon. He said, this suggests that in general current reporting of genotyping tests from labs may lack quality control and reliability.
The data Boucher collected and presented shows that of the labs he evaluated they were more successful at detecting mutations when 100% of a person's virus population were mutant than when 25% of a person's viral population were mutant. The lower the percentage of mutant viruses in a person's viral population the less successful the labs were at detecting the mutants. He said there were large differences in the sensitivity and accuracy of currently used genotyping methods.
The following table was displayed showing that when the percent of mutant viruses in a person's total population was low the number of mutants detected was also low, suggesting that a lab has more difficulty in detecting mutant viruses when they are present at a low level. A table with similar results was shown regarding the 215 codon. The question is are mutant viruses present at such a low level relevant to predicting resistance and non-response to a drug? The answer to that question is-we don't know. Dr Boucher suggested testing a person's DNA in cells for resistant mutant viruses is more reliable than testing the genotypes in RNA. He said labs can do this upon request. A mutant virus not detected in RNA in the blood or detected at a 10% level of presence may be more detectable in the DNA in cells. He said this may be particularly true in the case where a person stopped taking a drug years ago and resistance mutations are no longer detectable in RNA in blood. The person's archival mutants may still be detectable in DNA in cells. He thinks testing DNA is a more reliable way of detecting resistance.
Detection of Codon 41 Mutation
(qualitative)
% Mutant Input |
Mutant Detected |
No Mutant Detected |
Incorrect Codon Call |
|
# of Labs |
# of Labs |
# of Labs |
| 0 | 1 |
20 |
1 |
| 5 | 1 |
20 |
0 |
| 10 | 4 |
10 |
2 |
| 25 | 13 |
8 |
1 |
| 50 | 18 |
3 |
1 |
| 75 | 20 |
0 |
1 |
| 90 | 20 |
2 |
0 |
| 95 | 20 |
1 |
1 |
| 100 | 20 |
2 |
0 |
Boucher says this data indicates the level of unreliable results reported from labs conducting genotyping. Further, Dr Jonathan Schapiro, also speaking at this symposium, says that Boucher's research only represents the situation regarding mutations that apply to 2 drugs (AZT & 3TC) and only 1 class of antiretrovirals (NRTIs). Schapiro maintains, if you were to conduct similar research with other classes (PIs, NNRTIs) additional margins of error would emerge making detecting and reporting resistance increasingly more unreliable. It was suggested at this symposium that labs needed better training of personnel, less variability between labs (better standardization) and better regulation and certification of their quality control by an independent body. I've heard rumors prior to this symposium and at this symposium that the FDA may be requiring such regulation of the reliability of all genotyping and phenotyping testing by labs. I've tried to confirm this with the FDA but so far I've been unable to do so. If this is true, labs offering resistance tests such as Specialty Labs and LabCorp will be required show that their resistance testing is quality controlled and reliable.
Perhaps the doubts raised at this symposium sponsored by Visible Genetics (a manufacturer of genotyping testing) about the reliability of genotyping and phenotyping testing need confirmation from independent sources, but awareness of these doubts may be important. Many doctors and patients are using resistance test results in making treatment decisions. Are they asking the labs about the reliability of their testing procedures? Is there no FDA regulation of the performance of these tests? Do we need such regulation, and FDA approval as is the case with viral load testing? Which 23 labs were included in Boucher's evaluation? Did they include LabCorp who uses Virco's genotyping and phenoyping testing?If so, how did LabCorp perform. It may be premature to generalize Boucher's data unil more information is available.
Finally, Dr Clive Loveday of the Royal Free Hospital in England discussed his preliminary findings from a study treating 60 people who have been heaviy pretreated and failing therapy. Most of the patients are receiving a 5-drug regimen consisting of ritonavir+indinavir+efavirenz+hydroxyurea+ddI. But some individuals are receiving an additional 1 or 2 drugs; and some individuals who cannot tolerate ddI are receiving substitutes. Data for about 30 patients are available past 8 weeks. Sixty percent have below 400 copies/ml with followup out to 12 to 24 weeks. An abstract will be presented at the 1999 Human Retrovirus Confeence in late January. As well, Sunday at this Conference Dr Cassy Workman will present extended data on the combination of ritonavir+indinavir presented at Geneva.