Report 9 -  4th International Workshop on HIV Drug Resistance and Treatment Strategies
Written by Jules Levin
Sitges, Spain, June 12-16 2000

D4T Resistance and Resistance Testing. AZT Resistance Mutations Observed
Following D4T Treatment

BRIEF SUMMARY OF THIS REPORT

There are 3 reports below from the Resistance workshop each of which relates to d4T resistance. Two reports present the AZT mutations they found and the incidence of their occurrence following d4T treatment in both AZT naÔve and in NRTI experienced individuals. The risk factors of their occurrence are discussed. Prior to this research being presented at this year's Resistance Workshop I think it was generally agreed that treatment with an NRTI might reduce a person's virologic response to subsequent NRTI treatment.

One of the reports below which was presented by Vincent Calvez and a French research group at Sitges highlights one of the concerns surrounding resistance testing. That is, itís difficult to detect resistance to d4T using genotypic and phenotypic resistance testing. D4T resistance may be due to mutational patterns not yet completely identified, or maybe d4T resistance does not develop very easily, or possibly phenotypic testing cut-offs have to be revamped to be able to assess d4T resistance. What is a phenotypic"cut-off"? For the Virologic phenotypic test the cut-off between no resistance and intermediate resistance is >2.5. That means a 2.5 fold increase in the amount of drug needed to suppress HIV replication. This is often referred to as the IC50 (inhibitory concentration of drug needed to suppress 50% of HIV replication). A 2.5 fold increase means you need 2.5 more drug to inhibit 50% of HIV replication. A 2.5 fold increase refers to a 2.5 times increase in IC50. At the moment the Virco cut-off between sensitive and intermediate resistance is 4.0. So the two commercially available tests have different cut-offs (2.5 vs 4.0). This creates a situation where a value in between these 2 points of for example 3.0 will be reported as intermediate resistance by Virologic but sensitive by Virco. I think it was generally agreed at the Resistance Workshop that resistance tests need further refinement in a number of areas and this is one of them. The manufacturers of these tests said at the meeting that they will be working to address these types of questions over the next year or two. 

Resistance and Viral Response to d4T/3TC Combination in AZT, ddI, and ddC Experienced Patients in the Altis 2 ANRS Trial

The ALTIS study observed individuals who had previous NRTI experience and were switched to a d4T regimen. My recollection of the viral load data from the ALTIS study which was reported over 1 year ago was that participants did experience a reasonable median viral load reduction. Calvez intended this study to evaluate the factors associated with decrease in plasma HIV-RNA at week 24 in the ALTIS 2 Trial. By the way of background the study authors saidóamong the approved antiretrovirals, d4T has been the most difficult with ddI to demonstrate specific phenotype and genotype resistance in the clinic. Calvez concluded that the results from this resistance study showed a relative phenotypic and genotypic cross-resistance between d4T and AZT. Of interest is that he also concluded that modest increases of d4T IC50 and IC90 may change the viral load response showing the need for a new definition of decrease in d4T susceptibility.

RT sequences of plasma RNA and proviral DNA (ABI, Perkin-Elmer test) and phenotype (PBMC test) were performed at baseline of this study on 26 patient samples. RT sequences refers to identifying the genotypic changes on the reverse transacriptase enzyme that relate to NRTI resistance. Based on the results observed with 50 susceptible viruses, when d4T IC50 was >0.27uM, the virus was classified resistant, when d4T IC50 was £0.27 uM and d4T IC90 was >1.65 uM, the virus was classified intermediate, and susceptible in the remaining cases. The d4T index was expressed as the ratio of IC50 for the sample to that of the mean value (0.15 uM) observed in susceptible viruses. Non-parametric tests were used for univariate analysis and stepwise regression to determine independent prognostic factors of the virologic response among those for which univariate P-valu was lower than 0.20.

At baseline, most of the subjects had, in plasma and PBMCs, AZT associated mutations. At week 24 all patients acquired the 3TC mutation (M184V). The factors associated with decrease in plasma HIV-RNA were:

For d4T index threshold values higher than 1,8 (threshold of 4.0 is used in Virco test and 2.5 is used in Virologic phenotypic test) the viral response was very modest (decrease lower than 0.3 log copies/ml).

In the multi-variate analysis, the stepwise model with the higher multiple correlation coefficient (r2=0.707) included the presence of the 215Y or F mutation (p=0.0001), the number of previously used NRTIs (p=0.0210) and a resistant d4T phenotypic test result (p=0.0295).

Incidence of d4T Resistance in d4T Experienced Individuals and Risk Factors for Decreased Sensitivity to d4T

Another important point that emerged from the Workshop was that for those treatment-naÔve individuals taking d4T AZT resistance mutations can occur. Santiago Moreno, a Spanish research group, and Brendan Larder from Virco (abstract 23) reported on a group of 41 patients that had been treated with a median of 3.2 (2-5) NRTI drugs for a median of 45 (6-90) months. In addition they had received protease inhibitors (median 1.3) for 18.5 months. Combined genotypic (ABI sequencing) and phenotypic (recombinant virus assay) resistance testing of isolates from 41 patients after 2 or more virologic failures of d4T containing regimens were performed. Viral isolates were classified as susceptible (<4 fold increase of IC50 over wild-type), intermediate resistant (4-10 fold increase) and resistant (>10 fold).

Moreno reported that 73% (n=30) of the isolates remained fully susceptible, while 22% (n=9) were intermediately resistant, and only 5% (n=2) were resistant to d4T (11- and 65- fold increase). In univariate analysis, mean number of NRTIs previously used, and fold resistance to AZT, ddI and abacavir were associated with increased fold resistance to d4T. In a logistic regression analysis, only the mean number of previous NRTIs (2.41 versus 3.6, (p<0.001), and increased fold resistance to AZT (26.3 versus 94, p<0.001) were predictors of increased resistance to d4T. There was a linear correlation between increased fold resistance to AZT and d4T (R=0.74, p=0.001).

Moreno concluded that decreased susceptibility to d4T can be seen in up to 27% of patients treated with multiple NRTIs and after 2 or more failures with d4T. The number of NRTIs previously taken and the phenotypic resistance to AZT are associated with increased fold resistance to d4T.

AZT Mutations and Differing Mutational Patterns Observed in Firstline Regimens with d4T/ddI and d4T/3TC

Graeame Moyle and a Britsh research group reported finding in their study of 47 AZT-naÔve d4T treated patients that AZT type mutations were commonly observed. He also observed in this study that the choice of co-nucleoside might influence the mutational pattern seen. And, when ddI was used in combination with d4T no NRTI mutations were seen in over 20% of the patients. But, when 3TC was used with d4T the 3TC M184V mutation was almost always seen.

 The purpose of the study was to look at the type and frequency of different thymidine (AZT) mutations when d4T was given with either ddI or 3TC. Over 700 VircoGen genotypes have been performed at the Chelsea and Westminster Hospital in London. Moyle extracted data on RT mutations in patients receiving their first ever regimen containing d4T plus either ddI or 3TC.

Forty-seven patients were identified, 24 on ddI and 23 on 3TC. Thymidine-type mutations (at codons 41, 67, 70, 210, 215, and 219; these are AZT mutations) were observed in 68% (n=32) of the samples. The mean number of ddI RT mutations were 1.58 and there were 2.1 for 3TC treated patients, with no RT mutations present in 21% (n=5) of ddI and 4% (n=1) of 3TC treated patients. The M184V mutation was observed in 1 (4%) of ddI and 22 (96%) of 3TC treated patients, and was the sole mutation in 8 (35%) of 3TC treated patients.

The Q151M multi-nucleoside resistance mutation was observed in only one person in the ddI group. Mutations K65R (2 patients), I74V (2 patients, and V75T (1 patient) and 210W (4 patients) were only observed with ddI. Whereas M41L occurred at similar frequency (7 ddI and 6 3TC recipients), 215Y/F was more common with ddI ñ 11 (46%) versus 6 (26%) 3TC recipients (p=0.27)ówhereas K70R was more common with 3TC (1 ddI versus 6 3TC recxipients; p=0.89).