Data presented on Monday at the Fifteenth Conference on Retroviruses and Opportunistic Infections in Boston examined the effect of increasing doses of protease inhibitors based on therapeutic drug monitoring. While this strategy did not result in virologic gains for patients overall, it did result in modest gains in viral suppression in black and Hispanic patients, and in those with moderate degrees of resistance to protease inhibitors.
Drug effectiveness based on pharmacokinetics can be evaluated in a variety of ways
The inhibitory quotient (IQ) is calculated as the actual minimum drug concentration (Cmin) achieved, divided by the IC50, which is the concentration needed to inhibit viral replication by 50%. Thus, the IQ is a measure of how robustly the actual drug concentrations exceed the minimal required levels.
In this study (A5146), dose adjustment was determined using a modification of the IQ known as the normalised inhibitory quotient (NIQ). For each PI, a reference IQ is derived from a population that has been virologically controlled on that drug: hence, the reference IQ should represent the threshold for virologic success. The NIQ is the ratio between the IQ observed in each patient and the reference IQ for that patient's drug, i.e., IQpatient/IQreference. An NIQ ≤ 1 can represent a low drug trough concentration, significant drug resistance, or both.
This trial studied treatment-experienced adults who had failed at least one protease inhibitor and had viral loads ≥ 1000 copies/mL while on antiretroviral therapy. At study entry, patients began a new protease inhibitor (PI)-based regimen based on virtual phenotyping.
At week two on the new regimen, PI trough concentrations were assessed, and NIQs were calculated based on the trough concentrations.
At week four, those with NIQ > 1 continued their regimen unaltered. Those with an NIQ ≤ 1 were randomised to either continue their regimen unaltered, or to have their PI dose increased. The reported analysis was limited to the 183 patients with NIQ ≤ 1, 90% of whom were men, 49% white non-Hispanic, 25% black, 24% Hispanic, median age 45 years, and median number of active PIs 0.7 by vircoTYPE susceptibility score. Of these, 92 were randomised to dose adjustment, and 91 to continued standard of care. Median viral load was 4.36 log10 copies/mL at study entry and 3.64 log10 copies/mL at randomisation; median CD4 cell count was 194 cells/mm3 at study entry and 221 cells/mm3 at randomisation.
All FDA-approved antiretrovirals were allowed; darunavir was not used by any study participants as it had not yet received FDA approval. All PIs were ritonavir-boosted except for nelfinavir. Ritonavir-boosted dual PIs were used by 47% of the participants; fosamprenavir plus saquinavir in 18%, and Kaletra plus saquinavir in 17%. Thirteen percent began a drug from a new drug class; this was T-20 in most (78%) of these cases.
Median PI trough concentration and NIQ increased significantly more in the patients who received TDM-based dose adjustments, except for those (32% of the study patients) on fosamprenavir. The investigators were unable to explain the inability to increase fosamprenavir concentrations with dose escalation.
However, there was no significant difference in the primary outcome – change in viral load at 20 weeks after randomisation – between the strategies in the study population overall. At 20 weeks, viral load was essentially unchanged from the point of randomisation in both groups: 0.09 log10 in the control arm (95% confidence interval [CI], -0.17 – 0.66) and 0.02 log10 in the TDM arm (95% CI, -0.49 – 0.57) (p=0.17). There were also no significant differences in time to virologic failure, proportion with viral load < 50 copies/mL or < 400 copies/mL, or CD4 response. On-treatment analysis and analyses stratified by whether a new class of drug was added also failed to show a significant difference.
However, outcomes differed significantly by race. Outcomes were significantly improved, by approximately half a log, in black (p=.05) and Hispanic patients (p=.04) who received TDM-based dose escalation, while there was no significant difference in white patients (p=0.4). There was also a significantly greater reduction in viral load (slightly over half a log, p=.002) in patients who were using more than the median of 0.7 new active PIs; this difference was more pronounced in blacks and Hispanics. The racial differences were not explained by differences in body mass index, baseline viral load or CD4 cell count, drug resistance or adherence levels, or fosamprenavir use.
The investigators concluded that there was "no overall benefit of TDM in this study," but note that TDM "may confer more benefit in black and Hispanic patients than white patients, for unclear reasons." Post-hoc analyses "suggest a TDM benefit in patients whose virus has some ability against their PIs," leading the researchers to propose that increased drug levels might not overcome high levels of drug resistance, but that they might have utility when resistance levels are intermediate.
Demeter L et al. A prospective, randomized, controlled, open-label trial evaluating the effect of therapeutic drug monitoring and protease inhibitor dose escalation on viral load responses in antiretroviral-experienced, HIV-infected patients with a normalized inhibitory quotient. Fifteenth Conference on Retroviruses and Opportunistic Infections, Boston, abstract 35, 2008.