CROI: Etravirine (TMC125) and darunavir (TMC114) in treatment-experienced patients

Researchers from Tibotec Pharmaceuticals presented data on two experimental agents, the non-nucleoside reverse transcriptase inhibitor (NNRTI) etravirine (TMC125) and the protease inhibitor (PI) darunavir (TMC114), used in treatment-experienced patients, at the Thirteenth Conference on Retroviruses and Opportunistic Infections, held in Denver.

Johan Vingerhoets presented results from study TMC125-223, looking at the impact of baseline resistance on response to etravirine in patients with extensive NNRTI and PI resistance. Tibotec designed the new drug to have a high genetic barrier to the development of resistance and activity against NNRTI-resistant HIV.

Study TMC125-223 included 199 individuals with extensive prior antiretroviral therapy experience, documented NNRTI and PI resistance, and viral loads of at least 1,000 copies/ml. Participants were randomly assigned to receive one of two doses (400 or 800mg twice daily) of etravirine plus an optimised background regimen, or else an optimised control regimen of approved antiretroviral drugs.

After 24 weeks on therapy, using an intent-to-treat analysis, HIV viral load decreased by a mean 1.18 and 1.04 log₁₀ copies/ml in the 800mg and 400mg etravirine arms, respectively, compared with 0.19 log₁₀ copies/ml in the control group (p < 0.05). These data(data were presented at the European AIDS Conference in November 2005.

The current analysis focused on baseline resistance parameters associated with decreased susceptibility to etravirine. Almost all patients had previously used a NNRTI, and they had a median two NNRTI-resistance mutations (range 0-5), the most common being K103N, Y181C, and G190A. Patients had a median six nucleoside reverse transcriptase inhibitor (NRTI)-resistance mutations, at least three (median four) primary PI-resistance mutations, and a median nine PI-resistance-associated mutations; 83% showed no phenotypic susceptibility to any PI (excluding the recently approved tipranavir (Aptivus)).

The median baseline fold-change in EC₅₀ (50% maximum effective concentration) to etravirine was 1.7, compared with about 40 for efavirenz (Sustiva) and about 60 for nevirapine (Viramune). No single mutation alone produced a mean etravirine fold-change greater than 10. Seven mutations, including K101P, V179E, Y181I, Y181V, G190S, and M230L, were associated with fold-changes greater than 10, but only in combination with up to four additional mutations (usually including Y181C). Even a combination of mutations produced a fold-change greater than 10 in just 12% of patients.

Etravirine activity was greater in individuals with fewer baseline NNRTI-resistance mutations. After 24 weeks, among the 79 patients receiving the most effective dose (800mg twice daily), viral load decreased by 1.82, 1.65, 1.00, and 0.66 log₁₀ copies/ml in patients with 0, 1, 2, and three or more mutations, respectively. In patients with multiple mutations, virological response to etravirine still exceeded the 0.19 log₁₀ copies/ml decrease seen in the control arm.

The researchers concluded that etravirine “retains activity in the presence of multiple NNRTI mutations where current NNRTIs are not expected to be effective.”

M.P. De Béthune from Tibotec presented combined resistance data from the POWER1 and POWER2 studies of darunavir in heavily treatment-experienced patients.

Participants in the two Phase IIb studies were taking PI-containing regimens at study entry, had prior exposure to all three classes of oral antiretroviral drugs, had at least one primary PI-resistance mutation, and had viral loads of 1,000 or more copies/ml. They were randomly assigned to receive one of four doses of darunavir boosted with ritonavir (Norvir) or else a comparator PI selected by the investigators. All participants also received an optimised NRTI background regimen, and some took T-20 (enfuvirtide, Fuzeon).

Efficacy and safety data from POWER1 (were reported at the International AIDS Society conference in July 2005) and results from POWER2 were presented at the (Interscience Conference on Retroviruses and Opportunistic Infections in December 2005).

POWER1 and POWER2 included 318 and 278 participants, respectively. In this pooled subanalysis, the combined mean baseline viral load was 4.6 log₁₀ copies/ml; patients had a median of eight baseline PI-resistance mutations and a mean six years duration of prior PI use. (Participants in POWER2 actually had somewhat more advanced disease, with higher viral loads and lower CD4 cell counts). In the control arm, the most commonly used comparator PIs were lopinavir/ritonavir (Kaletra), saquinavir (Invirase), amprenavir (Agenerase) and atazanavir (Reyataz).

After 24 weeks, in an intent-to-treat analysis, viral load declined by a mean 1.89 log₁₀ copies/ml among the 131 participants taking 600mg darunavir/100mg ritonavir twice daily (the most effective dose), compared with 0.58 log₁₀ for patients taking active comparator PIs (n = 34) and 0.43 log₁₀ for those whose virus was resistant to their comparator PIs (n = 86). In the most effective darunavir arm, 70% of patients achieved at least a 1 log reduction in HIV RNA, compared with 32% and 16%, respectively, in the sensitive and resistant comparator PI arms. The corresponding proportions achieving viral loads below 50 copies/ml were 45%, 24%, and 7%. In a multivariate analysis, viral load reduction was significantly greater for boosted darunavir than for either sensitive or resistant comparator PIs (p < 0.0001).

The current subanalysis examined the number and type of baseline viral mutations and those that developed during POWER1, POWER2, and a third supportive trial, POWER3, representing a total of 458 patients who received the 600/100mg dose of darunavir/ritonavir. Participants were stratified based on whether or not they used T-20.

The researchers determined that fold-change in EC50 was the most important predictor of virological response to darunavir. At least one-third of patients with 1-9 baseline PI-resistance mutations achieved viral loads below 50 copies/ml, but the response rate dropped off for those with ten or more mutations.

Eleven baseline mutations, V11I, V32I, L33F, I47V, I50V, I54L, G73S, L76V, I84V, and L89V, were associated with diminished virological response to darunavir (defined as less than one-third of subjects achieving viral loads below 50 copies/ml), though the response rates in these patients still exceeded those seen in the comparator PI arm.

In terms of new mutations, an analysis of patients in the darunavir arm with virological failure (51 who showed initial virological response followed by rebound and 95 who never achieved at least a 1 log drop in HIV RNA) showed that 10% or more developed the mutations V32I, L33F, I47V, I54L, and/or L89V. Among the rebounders, there was a median 8.14 fold-change in darunavir activity from baseline; however, these patients did not show decreased susceptibility to tipranavir.

Constructed site-directed mutant viral strains harbouring V32I, L33F, I47V, or I54L alone or with 1-2 additional mutations did not demonstrate decreased susceptibility to darunavir, leading the researchers to suggest that a higher total number of mutations is required to confer resistance.

The researchers concluded that boosted darunavir “demonstrates a significant efficacy benefit” over comparator PIs in treatment-experienced patients, and that diminished response is associated with the number of baseline resistance mutations.

Based on the results of the POWER studies, Tibotec submitted a new drug application to the US. Food and Drug Administration in December 2005 and to the European Agency for the Evaluation of Medicinal Products in January.