Lead optimization activity cliffs

Although the magic methyl phenomenon is not rife in lead optimization data (if it were, the entire premise that a lead may be optimized by synthesis of close analogs would be discredited), recent studies on large datasets now reveal a consistent but low-probability effect of potency increases on the addition of a methyl group [19], the presence of activity cliffs in many biological response surfaces [20], and the extent to which assumptions of additively are valid [21] only 50% of the (single parameter) datasets studied exhibited additivity. 

Many instructive exceptions of the similarity principle were summarized by Kubinyi [48,49]. S ome of these differences could be related to unexpected 3D-bmding modes of ligands in the protein cavity after only minor chemical modifications. These examples show that it is very difficult to describe chemical similarity in an objective and global manner without considering the protein environment. In fact, lead optimization often takes advantage on these surprises and further explores such activity cliffs for biological activity. Regions around early hit structures with a flat SAR are typically less often explored. 

The systematic exploration of activity cliffs from various representations of activity landscapes is summarized in an interesting article by Stumpfe and Bajorath [79]. These representations could include either distinct chemical transformations (MMPs) or similarity-based relationships between molecules. In a classical medicinal chemistry approach, activity cliffs are extracted from R-group tables, as shown in Figure 10.3 for 3-oxybenzamides as factor Xa inhibitors [58, 59]. This has to be repeated for each scaffold of interest with a predefined view on informative attachment points at the molecular core. Hence this approach is feasible for smaller datasets with only a few informative attachment points for SAR investigation, typically a narrow series in lead optimization. 

---