• Slide 237: Phenotypic Resistance Testing Significantly Improves Response to Therapy (Tx): A Randomized Trial (VIRA 3001) (Authored by C. Cohen, S. Hunt, M. Sension, C. Farthing, M. Conant, S. Jacobson, J. Nadler, W. Verbiest, K. Hertogs, M. Ames, A. Rinehart, and N. Graham. Virco, Belgium; Glaxo Wellcome, NC)
  Click here to view the original abstract

• Slide 238: Selection of Antiretroviral Resistance in the Latent Reservoir of Human Immunodeficiency Virus Type 1 during Successful Therapy (Authored by J. Martinez-Picado, M.P. Depasquale, N.A. Kartsonis, G. Hanna, J. Wong, D. Finzi, E. Rosenberg, H. Gunthard, L. Sutton, A. Savara, C. Petropoulos, N. Hellmann, B.D. Walker, D.D. Richman, R. Siliciano, and R.T. D'aquila. Massachusetts Gen. Hosp. & Harvard Med. Sch., Boston, MA; Fndn. irsiCaixa, Barcelona, Spain; Univ. of California San Diego and VA Med. Ctr., La Jolla, CA; Johns Hopkins Univ., Baltimore, MD; and ViroLogic Inc., South San Francisco, CA)
  Click here to view the original abstract

Slide 237

Resistance testing has become more and more popular, although controlled trials demonstrating benefit of these tests has been scant. GART and VIRADAPT, despite problems with trial design, have shown that genotypic testing can be beneficial in choosing a new drug regimen after virologic failure. This is the first randomized trial to show a similar benefit with phenotypic resistance testing.

Subjects in this trial were failing (viral load >2000) their first regimen with a protease inhibitor and two nucleoside RT inhibitors. Patients were randomly assigned to receive a new treatment regimen based on either the current "standard of care" or phenotypic testing (with the Virco test); 107 and 111 patients began treatment in each group respectively. The median CD4 cell count was about 350, and mean viral load about 10,000, although with a range from 2000 to >750,000; nelfinavir was the most common PI in the original regimen, followed by indinavir. 3TC, AZT and d4T were the most common RT inhibitors.

Amprenavir and saquinavir showed the highest level of sensitivity, not surprising given the previous treatment history. Those treated with nelfinavir generally showed a higher level of sensitivity to other protease inhibitors than those treated with indinavir. Amongst the RT inhibitors 3TC had the highest level of resistance, d4T the least. Results of the phenotype did influence choice of drugs: 78% of patients had a change in the planned treatment regimen after the results were available.

A preliminary analysis is based on PCR <400 at week 16. Those in the phenotype arm had 38% "undetectable" versus 23% in the standard of care arm (intent to treat analysis), or an average PCR decrease of 1.27 vs. .75 log (observed data). This improved outcome was seen regardless of baseline viral load.

Although the analysis is preliminary, this trial gives us strong evidence that use of phenotypic resistance testing can significantly improve the outcome in the setting of a first protease failure. Whether the phenotypic assay provides better outcome than a genotypic test, particularly given the substantial increase in cost and time for results, remains to be seen.

Slide 238

It has long been known that, with sophisticated virologic techniques, HIV can be isolated from resting-memory CD4 T cells even in patients who are successfully suppressed to <50 copies/ml. Previous reports have not demonstrated development of mutations associated with resistance in this situation.

In this trial virus was isolated and cloned over multiple time points in patients on protease-based regimens with PCR <50. In many of these patients mutations associated with resistance developed progressively overtime. However these mutations were only seen in a minority of clones, and no patients became clinically resistant.

None of the patients who developed mutations were on their first treatment regimen, implying pre-existing resistance. Additionally, the development of mutations was significantly more common in those who had low-grade "blips" of viral activity.

Although no clinical resistance was seen in this study, it raises the concern of subclinical development of resistance even in patients who are well controlled. Given this data, maybe we should be considering "proactive" regimen switches to prevent future treatment failure.