The genotypic inhibitory quotient (GIQ) is the ratio between drug concentration and the number of resistance mutations. However, as different resistance scores can be calculated for the same protease inhibitor, the ability of a GIQ to predict the virological outcome may vary depending on the resistance score used as the denominator. Forty-four highly treatment-experienced HIV-infected patients failing on their current regimen were treated with a lopinavir/ritonavir-containing combination for at least 48 weeks. Three GIQs were calculated for each patient: the denominator of the first (GIQ-A) was the lopinavir/ritonavir score calculated using the mutations listed by IAS-USA as being related to lopinavir/ritonavir resistance; the denominator of the second (GIQ-B) was the total number of mutations related to resistance to any protease inhibitor as reported by IAS-USA; and the denominator of the third (GIQ-C) was the lopinavir/ritonavir score proposed by Parkin et al. The median (IQR) of the GIQ-A, B, and C was 4.69 (3.83-9.76), 0.97 (0.74-1.62), and 1.02 (0.68-2.52), respectively. At week 48, the median decrease in HIV-RNA was 1.09 (0.32-2.34) log(10) copies/ml (P < 0.0001), with 13 subjects (29.5%) attaining undetectable levels. All of the GIQs independently predicted the change in viral load from baseline and undetectable HIV-RNA at week 48. The partial R(2) of GIQ-C was greater than that of GIQ-B, which was greater than that of GIQ-A. All of the GIQs were independent predictors of the 48-week virological response. The predictive value of the GIQ for lopinavir/ritonavir may vary depending on the algorithm used to score drug resistance.
(c) 2006 Wiley-Liss, Inc.