 |
|
|
| |
Cumulative Resistance Score Gives Broader View of Potential Antiretroviral Activity
|
| |
| |
7th European HIV Drug Resistance Workshop, March 25-27, 2009, Stockholm
Mark Mascolini
A resistance score incorporating all mutations ever detected in a person's virus--not just mutations in the last genotype--afforded a deeper look at potentially inactive antiretrovirals in a two-center study [1]. The results confirm and extended earlier findings in a Canadian cohort [2].
Federico Garcia and colleagues in Granada and London studied 227 people who had two or more genotypic resistance tests while taking antiretrovirals. They figured two genotypic susceptibility scores (GSS)--one based only on the latest genotype and one based on every genotype for each patient. Using the Stanford, ANRS, and Rega genotypic resistance interpretive systems, the investigators scored individual antiretrovirals as 1 (susceptible), 0.5 (intermediate resistance), or 0 (high-level resistance).
The study group included 158 men (70%) and 66 people (29%) with an HIV-1 subtype other than B. Median age stood at 43.7 years (range 13 to 74). These people had a median of two genotypes (range 2 to 9) from 1999 through 2008. They had tried a median of 6 regimens (range 1 to 21), and median time between first and last genotype measured 21.9 months (range 2 to 76).
Garcia and coworkers found an inaccurately higher overall median GSS with the most recent score (2.5, range 0 to 4) than with the cumulative score (2, range 0 to 4) (P < 0.05). The difference between cumulative and latest score was greatest in the nucleoside class because of the many M184V mutations counted in the cumulative score but not recorded in the latest score. With the three different genotyping systems, an interpretation change from susceptible to intermediate or high-level resistance occurred with nucleosides in 17% of cases, with nonnucleosides in 14% of cases, and with protease inhibitors in 5% of cases.
All three genotypic scoring systems found that the change from most recent to cumulative score had the least impact on etravirine among nonnucleosides, and on tipranavir/ritonavir and darunavir/ritonavir among protease inhibitors. In the protease inhibitor class, predicted change in resistance according to the three systems was greatest with saquinavir/ritonavir and indinavir/ritonavir.
An earlier study of 1734 antiretroviral-treated people in British Columbia, Canada, also found that the most recent genotype underestimated cumulative resistance, particularly with nucleosides [2]. But the Canadian study did not attempt to determine whether the cumulative resistance score provided a better response predictor. A 95-person study found that calculating a genotypic inhibitor quotient (GIQ, trough concentration/number of mutations) with cumulative protease mutations did yield a better GIQ association with virologic response to a lopinavir regimen [3].
Garcia and colleagues analyzed virologic outcomes in 76 people who had follow-up data after their last genotype. The investigators calculated viral load change from baseline to three points--12 weeks, 13 to 24 weeks, and more than 24 weeks. Then they compared results between people whose new regimen would be correct according to the cumulative score and people whose new regimen would be incorrect. These 76 people always had a superior virologic response if the regimen change was correct according to the cumulative score, though the superior response reached statistical significance only at the 12-week point (-0.9 versus -0.4 log copies/mL, P = 0.035).
References
1. Alvarez M, Garcia A, Guillot V, et al. Cumulative versus last genotype: Impact on the prediction of antiretroviral drug-resistance. 7th European HIV Drug Resistance Workshop, March 25-27, 2009, Stockholm. Abstract 4.
2. Harrigan PR, Wynhoven B, Brumme ZL, et al. HIV-1 drug resistance: degree of underestimation by a cross-sectional versus a longitudinal testing approach. J Infect Dis. 2005;191:1325-1330.
3. Hoefnagel JG, van der Lee MJ, Koopmans PP, et al. The genotypic inhibitory quotient and the (cumulative) number of mutations predict the response to lopinavir therapy. AIDS. 2006;20:1069-1071.
|
| |
|
|
|
|
|
