icon_folder.gif   Conference Reports for NATAP  
  ICAAC 41st Interscience Conference on Antimicrobial Agents and Chemotherapy
Chicago, Illinois, December 16-19
Back grey_arrow_rt.gif
Abacavir Resistance
Written by Jules Levin
  Randy Lanier and GlaxoSmithKline reported results from the ZORRO study at this year's ICAAC meeting, Dec 16-19 in Chicago (abstract 1761). ZORRO is a prospective, open-label, multicenter study designed to evaluatethe safety and tolerability of abacavir when used in combination with other drugs for both treatment-naive and experienced patients for a 16-week period of time when used in a clinical setting. This substudy looks at NRTI and abacavir resistance as determined by a virtual phenotype test prevalent in the patients with treatment-experience, and the factors associated with the presence of resistance.Due to the limitations of the testing methodolgy, all therapy experienced subjects who had virtual phenotyping must have had a viral load >1000 copies/ml prior to testing. Investigators used the Virco software program which has been developed to link genotypes to phenotypes, generating a "virtual" phenotype (vPT) from the genotypic resistance mutations. ZORRO is completely enrolled. A total of 83 3 patients were enrolled at 266 sites in the US. All patients were treated with abacavir (ABC) 300 mg twice daily and at least 2 other ART drugs. ABC could be substituted for a PI or NNRTI, but use of ABC as a single drug added to a failing regimen was not permitted. Of the 510 treatment-experienced patients in the study, 179 agreed to have virtual phenotyping. 37% of patients analyzed upon entering the study with viral failure had no NRTI resistance detected, using genotype with vPT. The results from these patients were presented in this study.
Baseline characteristics for 135 patients in the study were available. The average viral load was about 25,000 copies/ml and CD4 was 292. 75% were white, 38% Black, and 18% Hispanic. 15% female.
Lanier reported that in vitro studies have shown the following resistance mutations result from ABC: ABC initially causes the 184 (3TC mutation), and subsequently for combinations of 65R, 74V, 115F. In general 3 mutations are required for >8 fold resistance to ABC. Mutations selected by prior NRTI therapy conferring cross-resistance to ABC: 184 + 3 TAMS (NRTI mutations) (TAMS= 41L, 70R, 210W, 215Y/F, 219Q). As well, mutations associated with multidrug resistance can confer cross-resistance to ABC: Q151M, 69SS/insertion.
To the best of my recollection previous studies showed 3 or more AZT mutations could cause diminished response to ABC, with response decreasing with an increasing number of AZT mutations. This study found otherwise. Lanier reported no ABC resistance was found no matter how many TAMS were present unless 3TC resistance was also present. This suggests to me that additional research may be needed to address this conflict.
Lanier presented data previously reported from a meta-analysis of 5 GSK studies where ABC was added as a single agent to therapy. The data found that:
  • < 4-fold ABC phenotypic resistance did not appear to result in any resistance to ABC or it was very minimal so it should not affect response to ABC.

  • 4 to 7-fold resistance to ABC results in a modestly reduced response to ABC in perhaps 60% of patients. With some of these patients able to reduce viral looad only as much as 0.5 log. I think currently the Virco assay uses 4.5 as the cutoff for ABC resistance.

  • Patients with > 7-fold ABC resistance are very unlikely to respond to ABC and probably should not use it.

  • When there were no TAMS present, the presence of 3TC resistance did not result in ABC resistance.

In this study of 179 therapy experienced patients 90% had <4-fold ABC resistance, 9% had 4-7-fold resistance and 1% had >7-fold resistance. 29% had > 4-fold ABC resistance upon entering the study.
Lanier reported that in this study that when TAMS (NRTI mutations) were present without 3TC resistance (184) this never led to >4-fold ABC resistance. TAMS are listed above earlier in this article.
However, when 3TC resistance was present with 3 or 4 TAMS 4 fold resistance to ABC can occur. Resistance to ABC increases with the number of TAMS present when 3TC resistance is present. But according to the data in this study ABC resistance may only reach clinical significance when 3 or more TAMS are present.
2 TAMS, but not 1, resulted in AZT resistance of 5.5 fold. 3 TAMS resulted in 29-fold AZT resistance.
It appears that 3TC resistance may reduce the amount of AZT resistance for some patients for some period of time. But the median AZT resistance was still 10-fold when 3 TAMS+3TC resistance were present. 2 TAMS+3TC resistance did resulted in 3.1-fold ABC resistance.
The Virco assay used >1.7-fold resistance as the cutoff for d4T resistance. And 11% of the patients in this study had >1.7-fold d4T resistance. In general, patients with 2 or less TAMS did not have d4T resistance. Patients with 3 or more TAMS had >1.7-fold d4T resistance. When the 3TC mutation was present with 5 TAMS 2 patients had 5 fold d4T resistance. The presence of 3TC resistance appeared to reduce d4T resistance a little.
A report by Fenner (abstract I-1758) at ICAAC found that the V118I mutation was associated with ABC resistance.