Based on the results of this study it is recommended to reduce the dose of clazosentan to half in patients with moderate and to sellckchem one fourth in patients with severe liver impairment. Considering that clazosentan exhibits dose-proportional PK across the clinical dose range [11, 12], the results of this study can be applied to 5 mg h?1 and 15 mg h?1, the doses used to investigate the efficacy of clazosentan in the ongoing phase 3 studies. The reduction in systemic clearance and prolongation of elimination half-life in subjects with liver impairment is consistent with reduced excretion of the drug by the liver in these subjects. The reduced liver capacity in cirrhotic subjects to eliminate drugs which are excreted almost un-metabolized through bile could be the result of multiple factors, such as impairment of active uptake of the drug into hepatocytes, efflux of the drug into the bile duct or impaired blood perfusion of hepatocytes [20].

True steady-state conditions were not reached during the infusion in all subjects. A possible explanation could be the increase in half-life of clazosentan in subjects with liver impairment due to reduced drug clearance in these subjects, which delayed the attainment of steady-state conditions. Modelling and simulation were used to predict the concentration�Ctime profiles and PK parameters after longer infusion, i.e. a 24 h constant infusion of 1 mg h?1 in groups A, B, D and 0.5 mg h?1 in group C. Under these conditions steady-state was reached in all subjects.

It was shown that the increase in exposure to clazosentan in subjects with mild, moderate and severe liver impairment was similar after 24 h and after 6 h continuous infusion. This indicates that the results of this study can also be applied to longer infusion times. There was an increase in clazosentan exposure with increasing Child-Pugh score over the range of 5 to 13. The Child-Pugh score and three laboratory components of the Child-Pugh score, bilirubin, albumin and PT each showed a significant correlation with AUC(0,��), when these parameters were evaluated separately by linear regression. This suggests that the Child-Pugh score could be considered as a predictive marker of elevated exposure to clazosentan and the three laboratory components of the Child-Pugh score each contribute to this correlation.

Since clazosentan is highly protein bound (approximately 98%), a difference in protein binding was also explored between the liver impairment groups and healthy subjects. Mean percentage of unbound clazosentan in plasma was higher in the moderate and severe liver impairment groups (groups B and C) compared AV-951 with healthy subjects (group D). It has been reported that only for those drugs administered intravenously and which have a high hepatic extraction ratio and high volume of distribution, changes in protein binding might influence the PK of the compound.