| |
Unexpected drug-drug interaction between tipranavir/ritonavir and enfuvirtide
|
| |
| |
[RESEARCH LETTERS]
AIDS: Volume 20(15) 3 October 2006 p 1977-1979
de Requena, Daniel Gonzalez; Calcagno, Andrea; Bonora, Stefano; Ladetto, Laura; D'Avolio, Antonio; Sciandra, Mauro; Siccardi, Marco; Bargiacchi, Olivia; Sinicco, Alessandro; Di Perri, Giovanni
Department of Infectious Diseases, University of Turin, Turin, Italy.
Abstract
Fifty-five patients placed on tipranavir/ritonavir 500/200 mg twice a day (27 with enfuvirtide and 28 without) underwent tipranavir and ritonavir plasma concentration measurements by high-pressure liquid chromatography. Markedly higher tipranavir and ritonavir trough concentrations were observed in enfuvirtide recipients. The modelling of sparse plasma samples using a first order absorption and elimination monocompartmental model without time lag predicted higher tipranavir elimination half-life and volume of distribution in enfuvirtide takers. This unexpected drug-drug interaction warrants further investigation.
The association of enfuvirtide and tipranavir/ritonavir (TPV/RTV) has been shown to be effective as a core of salvage HAART [1], and has become a primary option for multi-experienced patients. Even if no drug-drug interactions are expected between TPV/RTV and enfuvirtide, no data have yet been obtained. In our unit, where therapeutic drug monitoring is done on a regular basis, tipranavir, ritonavir and enfuvirtide plasma levels were measured in a cohort of patients enrolled in the Tipranavir Expanded Access Programme.
Patients administered with TPV/RTV (500/200 mg twice a day) plus two nucleoside reverse transcriptase inhibitors with or without enfuvirtide (90 mg subcutaneously twice a day) were considered. Subjects not taking concomitant interacting drugs and with self-reported compliance of more than 90% in the past week were evaluated. Plasma samples were obtained at scheduled follow-up visits, and tipranavir and ritonavir plasma concentrations were measured using a validated high-pressure liquid chromatography method with ultraviolet detection. Samples obtained between 11 and 13 h after the last TPV/RTV dose intake were considered as a trough concentration (Ctrough), and comparison according to the concomitant administration of enfuvirtide was performed by using individual mean values. Modelling of sparse plasma samples was performed by using a first order absorption and elimination monocompartmental model without time lag. Time-averaged plasma tipranavir and ritonavir concentrations from each patient were modelled as naive pooled data according to enfuvirtide administration. Initial estimates of the volume of distribution (Vd = 10 l), constant of absorption (Ka = 0.49/h), and constant of elimination (Ke = 0.1/h) were obtained from a previous population pharmacokinetic study [2]. Parameter boundaries were allowed to be estimated by the software WinNonLin. This software was used for the modelling and estimation of pharmacokinetic parameters. Finally, tipranavir Ctrough was sequentially measured in subjects in whom enfuvirtide was either discontinuated or added to a tipranavir-based regimen. Student's t-test was used to study the differences between groups. Values were given as ng/ml.
A total of 463 samples from 55 subjects (27 with enfuvirtide, group A, and 28 without, group B) were considered. No differences in sex (male: 81.4% group A versus 82.1% group B, P = 0.4), weight (69 versus 70 kg, P = 0.8), height (175 versus 175 cm, P = 0.73), and hepatitis C virus co-infection status (18.5 versus 10.7%, P = 0.54), were observed between groups.
A total of 194 Ctrough samples were considered (105 from group A). The mean (± SD) tipranavir and ritonavir Ctrough concentrations were 34 431 ng/ml (± 20 010) and 279 ng/ml (± 269), respectively. A higher mean tipranavir Ctrough was observed in group A (41 069 ± 20 174 ng/ml versus 27 261 ± 17 516 ng/ml, P = 0.011), as well as a higher mean ritonavir Ctrough (371 ± 333 ng/ml versus 188 ± 139 ng/ml, P = 0.017). The mean (± SD) enfuvirtide Ctrough concentration in subjects in group A was 3610 (± 1399). No correlation between enfuvirtide and either tipranavir or ritonavir Ctrough values was detected.
The modelling of individual averaged tipranavir plasma concentrations gave a correlation coefficient (R) of 0.49 for group A and 0.615 for group B. A higher volume of distribution (Vd/F) (9.85 versus 6.5 l) and lower constant of elimination (Ke) value (0.07 versus 0.13/h) were observed in group A compared with group B, whereas the constant of absorption (Ka) was similar in both groups (0.66 versus 0.65/h). Half-life elimination (t1/2) was 9.69 and 5.36 h in groups A and B, respectively. A higher minimum concentration was predicted in group A (41 812 versus 26 668 ng/ml), without marked differences in the maximum concentration (71 955 versus 67 779 ng/ml) and area under the concentration time curve (709 481 versus 595 490 ng/h per millilitre). The modelling of ritonavir concentrations gave R = 0.35 for group A, and R = 0.47 for group B. Results similar to those for tipranavir were observed for ritonavir, the differences being more pronounced in Vd/F (115.8 versus 95.5 l) and Ke (0.21 versus 0.34/h). Differences were also observed in the half-life (3.17 h in A versus 2 h in B) and in the predicted minimum concentration (385 ng/ml in A versus 189 ng/ml in B), whereas no appreciable differences were predicted either for the maximum concentration (849 versus 859 ng/ml), or for the area under the concentration time curve (7970 versus 6630 ng/h per millilitre).
In two subjects who discontinued enfuvirtide, tipranavir Ctrough concentrations decreased by 50.8 and 25.6% (28 726 to 14 133 and 55 979 to 41 650 ng/ml), respectively. On the other hand, in the only subject who added enfuvirtide to a tipranavir-based regimen, the tipranavir Ctrough increased from 29 045 to 50 051 ng/ml (72.3%).
To the best of our knowledge this is the first report showing significantly higher tipranavir and ritonavir Ctrough in patients administered with enfuvirtide compared with values observed in individuals with no concomitant enfuvirtide intake (median values are reported in Fig. 1). This difference of tipranavir exposure was not attributable to an imbalance of weight, sex or liver disease between the two groups. Moreover, this trend was confirmed in two TPV/RTV takers who discontinuated enfuvirtide and one who added enfuvirtide to an ongoing regimen with TPV/RTV.
The putative mechanism of such an unexpected interaction is unknown. In-vitro studies with hepatic microsomes showed that CYP 3A4 is the predominant cytochrome P450 isoenzyme involved in tipranavir metabolism, whereas enfuvirtide is not a substrate for this metabolic pathway [3]. No clinically significant interaction was found between the latter and drugs with a high affinity for CYP3A4 [4,5]. On the other hand, enfuvirtide was found to have no significant inhibitory effect on the metabolism of probe drugs mediated by CYP3A4, CYP2D6 or N-acetyltransferase (< 20%), whereas a minimal inhibitory effect on CYP1A2, CYP2E1 or CYP2C19 was observed (≦ 30%) [6]. The latter, however, have not been reported to be significantly involved in TPV/RTV metabolism.
Pharmacokinetic modelling suggested that Vd/F and half-life of both tipranavir and ritonavir could be the parameters principally affected. Therefore, an elucidation of the mechanism of this interaction needs further evaluation.
The increase of approximately 50% in tipranavir exposure associated with enfuvirtide co-administration is not negligible and could theoretically be important from a clinical viewpoint. The efficacy of a tipranavir-containing salvage regimen has been shown to depend on the drug plasma concentration, as witnessed by the predictive role of the inhibitory quotient on the virological response in RESIST trials [2]. On the other hand, hepatotoxicity related to the administration of TPV/RTV has also been reported to be concentration dependent. In the same trials, grade 3 and 4 of hepatic toxicity occurred mostly in subjects with very high plasma concentrations [7]. Therefore, the clinical impact of such a drug-drug interaction warrants further evaluation.
References
1. Valdez H, McCalister S, Kohlbrenner V, Mayers D. Tipranavir/ritonavir (TPV/r) drives viral load (VL) below 400 copies/ml when combined with a second active drug (T-20) in protease inhibitor experienced HIV+ patients. In: 3rd IAS Conference on HIV Pathogenesis and Treatment. Rio de Janeiro, 2005 [Abstract WeOa0205].
[Context Link]
2. Yong C, Sabo L, Oksala C, MacGregor T, Kohlbrenner V, McCallister S, et al. Population pharmacokinetic assessment of systemic steady-state tipranavir concentrations for adults administered tipranavir/ritonavir 500/200 mg twice daily. In: 15th Conference of Retroviruses and Opportunistic Infections. Boston, 2005 [Abstract 654].
[Context Link]
3. Patel I, Zhang X, Nieforth K, Salgo M, Buss N. Pharmacokinetics, pharmacodynamics and drug interaction potential of enfuvirtide. Clin Pharmacokinet 2005; 44:175-186.
[Context Link]
4. Ruxrungtham K, Boyd M, Bellibas S, Zhang X, Dorr A, Kolis S, et al. Lack of interaction between enfuvirtide and ritonavir or ritonavir-boosted saquinavir in HIV-1-infected patients. J Clin Pharmacol 2004; 44:793-803.
[Context Link]
5. Boyd M, Zhang X, Dorr A, Ruxrungtham K, Kolis S, Nieforth K, et al. Lack of enzyme-inducing effect of rifampicin on the pharmacokinetics of enfuvirtide. J Clin Pharmacol 2003; 43:1382-1391.
[Context Link]
6. Zhang X, Lalezari J, Badley A, Dorr A, Kolis S, Kinchelow T, et al. Assessment of drug-drug interaction potential of enfuvirtide in human immunodeficiency virus type 1-infected patients. Clin Pharmacol Ther 2004; 75:558-568.
[Context Link]
7. Boehringer Ingelheim Pharmaceuticals, Inc. Tipranavir Anti-Viral Drugs Advisory Committee (AVDAC) briefing document. 19 April 2005. Available at: www.fda.gov/ohrms/dockets/ac/05/briefing/2005-4139b1-02-boehringer.pdf . Accessed: 25 May 2006
[Context Link]
|
|
| |
| |
|
|
|
