• Efficacy of Ritonavir/Amprenavir-Containing Regimen in HIV-1 Protease Inhibitor-Experienced Patients and Predictivity of the Delta Viral Load Values Up to Week 24 Using Genotypic Inhibitory Quotient (Oral 165)
  Authored by A.G. Marcelin, C. Lamotte, H. Ait Mohand, C. Delaugerre, N. Ktorza, P. Bossi, F. Bricaire, D. Costagliola, C. Katlama, G. Peytavin, V. Calvez

One of the difficulties with genotypic resistance testing is that although it tells you which mutations are currently present, it does not specifically tell you whether or not your patient will have a virologic response. Hence, many advocate for the use of both genotype and phenotype but again we have seen limitations both to the interpretation and success of regimens chosen based on those results. Therapeutic drug monitoring (TDM) has been touted as another useful tool for monitoring the actual drug levels in a patient and adjusting doses accordingly. Investigators in Paris evaluated whether an equation taking into consideration both the number of mutations and drug level could be formulated to predict who would respond to a boosted-amprenavir (APV, Agenerase) regimen.

Their objective was to evaluate the efficacy of ritonavir (RTV, Norvir) plus an amprenavir-containing regimen in HIV-1 protease inhibitor (PI)-experienced subjects, and to evaluate the predictive value of a genotypic inhibitory quotient (GIQ) on the virologic response at week 24. Forty-nine PI-experienced but amprenavir-naive subjects experiencing virologic failure were treated with ritonavir (100 mg BID) plus amprenavir (600 mg BID).

The genotypic resistance testing at baseline revealed a median of two major and four minor PI-resistance mutations. Subjects responded to therapy with a median HIV viral load (VL) decrease of -1.32 log by week 12 (W12) and -1.46 by week 24 (by intent-to-treat analysis). Baseline PI-resistance mutations (L10F/I/V, K20M/R, E35D, R41K, I54V, L63P, V82A/F/T/S, I84V) identified in univariate analysis and included in a genotypic score and amprenavir Cmin (a pharmacokinetic parameter used to evaluate therapeutic agents; the Cmin is the minimum level of the drug in someone's blood, Cmax is the peak value) at week eight were predictive of the virologic response at week 12. GIQ, calculated as the ratio of amprenavir Cmin to the number of HIV-1 protease mutations, was a better predictor of the magnitude of VL decrease at week 12 and week 24 than virologic or pharmacological variables used alone. Among patients with a detectable VL (more than 200 copies/mL) at week 24, the patients who failed without any further acquisition of PI-resistance mutations harbored an amprenavir week 24 Cmin higher than those with new PI mutations who failed.

The investigators concluded that the ritonavir plus amprenavir-containing regimen in this study shows a sustained efficacy in PI-experienced patients up to week 24. They suggest that GIQ could be used in therapeutic drug monitoring. This approach could help to define the plasma concentrations needed to control the replication of viruses with different levels of PI resistance (measured by the number of PI resistance mutations).

I think this is a novel approach but one that needs further study. How does one take into account genotypic variability that is dependent upon time of exposure to the various PIs as well as problems associated with the collection of a true Cmin? I personally am not convinced this offers any further benefit. For this methodology, one has to obtain the Cmin at week eight and results are not typically available for a week or more. By that time, one is approaching week 12 and if the patient has not had a virologic response by then, most likely his/her regimen would be changed. One should also consider the cost involved in this approach. It remains unclear to me whether there would be any cost or therapeutic benefit.