September 28, 2006
The introduction of highly active antiretroviral therapy has decreased morbidity and mortality and lengthened the lifespan of those living with HIV. In conjunction, increasing numbers of patients are classified as highly treatment experienced, and more highly drug-resistant HIV strains have arisen.
Knowledge of resistance patterns, especially those associated with cross-resistance between drugs within a class, is important for designing effective treatment regimens. In particular, drugs that exhibit minimal cross-resistance represent crucial components of clinicians' therapeutic armamentarium for highly treatment-experienced patients.
In this study, Peter Piliero and Douglas Mayers from Boehringer Ingelheim and Neil Parkin from Monogram Biosciences, Inc. (Monogram) have furthered our understanding of one such drug, tipranavir (TPV, Aptivus), and its role in highly treatment-experienced individuals.1
Tipranavir is a non-peptidic protease inhibitor (PI) with a unique cross-resistance profile that retains activity against the majority of multi-PI-resistant HIV-1 strains. When co-administered with ritonavir (RTV, Norvir), it is highly effective as part of a regimen for highly treatment-experienced patients and those infected with multi-PI-resistant HIV-1 strains.
Four key protease mutations at codons 33, 82, 84 and 90 are associated with PI cross-resistance. Phase II trials with ritonavir-boosted tipranavir demonstrated that most isolates with two or fewer of these key mutations remained susceptible to tipranavir and responsive to tipranavir/ritonavir-containing therapy.2 The purpose of Piliero et al's study was to evaluate the prevalence and impact of these key mutations on the susceptibility of available ritonavir-boosted PIs.
From January 2000 through February 2006, all samples received by Monogram for routine drug resistance phenotype and genotype testing were analyzed and included for analysis if they contained at least one drug-selected protease mutation (i.e., any non-wild-type amino acid at position 23, 24, 30, 32, 46, 47, 48, 50, 54, 82 (except I), 84, 88, 90 or L33I/F).
Recently updated clinical cutoff data were used to evaluate the proportions of samples susceptible to different ritonavir-boosted PIs by examining the L33F/I, V82A, I84V and L90M mutations, both individually and in combination. Fisher's exact test was used to examine the proportion of samples susceptible to tipranavir versus the other ritonavir-boosted PIs. Information on susceptibility to lopinavir/ritonavir (LPV/r, Kaletra), amprenavir (APV, Agenerase) and saquinavir (SQV, Fortovase, Invirase) was obtained from 9,860 isolates, susceptibility to atazanavir (ATV, Reyataz) was obtained from 9,007 isolates, and susceptibility to tipranavir was determined from 2,272 isolates. Darunavir was not included in the analysis, as it was still an investigational drug when this study was conducted.
The prevalence of key mutations is shown in the table below. Almost two thirds (62.4%) of the isolates had one or two key mutations, whereas more than one third (38.5%) had two or three key mutations.
Isolates with no key mutations were equally susceptible to all studied PIs.
As the number of key mutations increased, susceptibility to all the PIs steadily decreased. For example, at least half of the isolates were resistant to amprenavir (50.4%), atazanavir (57.8%) and saquinavir (69.6%) in the presence of two or more mutations.
With three or more mutations, the proportion of resistant isolates jumped to at least 70% in the presence of lopinavir (69.6%), amprenavir (85.1%), atazanavir (78.9%) and saquinavir (82.1%).
The majority of isolates (69.4%) exhibited resistance to tipranavir only when they contained all four of the key mutations.
In all, the L33F/I, V82A, I84 and L90M mutations affected tipranavir susceptibility to a lesser degree than they did the other ritonavir-boosted PIs. For isolates with up to three mutations, a significantly greater proportion remained susceptible to tipranavir versus ritonavir-boosted lopinavir, amprenavir, atazanavir or saquinavir (P < .0001). For isolates containing all four key mutations, a significantly greater percentage of isolates still remained susceptible to tipranavir as opposed to ritonavir-boosted amprenavir, atazanavir or saquinavir (P < .002).
Additional analysis revealed that the key mutations produced a differential impact on susceptibility of the isolates to different PIs. The I84V mutation, with or without the other key mutations, had the greatest impact on PI susceptibility, including tipranavir. In contrast, presence of the V82A mutation alone had no effect on tipranavir susceptibility, while it significantly impacted susceptibility to the other PIs. This is consistent with the fact that the V82A mutation is selected by ritonavir, saquinavir, lopinavir, and atazanavir, whereas tipranavir instead selects for the V82L and V82T mutations. L90M, which is associated with resistance to all PIs, had a very limited impact on tipranavir susceptibility. Overall, tipranavir was the least affected by any individual or combination of mutations in comparison with the other PIs.
Given that this study is a retrospective database analysis, a number of limitations should be highlighted.
Despite these limitations, the above findings support the use of tipranavir in highly treatment-experienced patients containing HIV-1 resistant to more than one PI. In phase II and III clinical trials, HIV-1 with two or more mutations at protease codons 33, 82, 84 and 90 has been shown to correlate with reduced susceptibility to tipranavir.3 These data were used to derive a unique tipranavir mutation score containing 21 mutations at 16 codons that affect susceptibility and virologic response.4 As an additional aid to the clinical management of PI-resistant individuals, the tipranavir package insert contains a table detailing the association between the number of specific tipranavir mutations and phenotypic fold-change ranges, as shown below.5
Since quantitative differences existed with regard to the specific combinations of mutations present, phenotypic testing may provide clinicians with additional guidance for making treatment decisions. This study supports the current guidance offered by the tipranavir package insert and provides additional detail on the differential impact of various combinations of key PI mutations.