Literature and Aventis Examples on Aspects of Multidimensional Optimization

Literature and Aventis Examples on Aspects of Multidimensional Optimization 

Sugano et al. studied the membrane permeation of 51 benzamidine-based thrombin inhibitors in a rat everted sac permeability model [197]. They reported significant membrane permeabilities in this in vitro model, which they attributed to passive paracellular transport, a different absorption mechanism to transcellular permeability. 

On the basis of their evaluation and our internal predictive VolSurf model [160] for this series (r 0.81, q 0.60, 4 PLS components), it can be concluded that factors like size and shape, which had previously been reported to affect paracellular permeability, are indeed important in the VolSurf PLS model to explain the local structure-permeability relationship of one particular scaffold. Hence, local statistical models provide a qualitative ranking of candidates, and thus are valuable for optimization of pharmaceutically relevant compounds, especially if combined with additional models to understand affinity, selectivity or any particular pharmacokinetic behavior. 

In the search for bUe-add resorption inhibitors (BARI), a predictive 3-D-QSAR pharmacophore model for the deal Na+/bile acid cotransporter was derived, which enhanced the understanding of binding and transport properties [205]. This model was then also successfully explored to search for potential substitution sites, which are not relevant for the SAR of this series, while they allow the addition of additional substituents to minimize the oral uptake of inhibitors. 

The approach discussed to use VolSurf derived in silica models to understand structure-PK relationships for pharmacokinetic properties was also applied to one series of selective cardiac KATP channel blockers [160]. It was found that compounds fulfilling the predefined selectivity profile exhibit only less-optimal pharmacokinetic properties because of a short plasma mean residential time (MRT). Consequently, the MRT for 28 compounds from rabbit iv studies for one series was used as dependent variable to derive a VolSurf PLS model in addition to ligand affinity SAR data. The chemical 

---