The mitogen-activated protein kinase (MAPK) signalling cascade is one of the key pathways regulating cell proliferation and differentiation and aberrant activation of this pathway is implicated in a number of cancers. Targeting components of the cascade thus has the potential to provide therapeutic agents with broad antiproliferative utility, including antitumour activity. Allosteric inhibitors of MEK, in particular, have been the focus of much activity in the industry and a number of compounds (e.g. PD0325901 and AZD-6244) have progressed to clinical studies, with AZD-6244 currently most advanced and the subject of numerous trials.Two recent letters describe the identification of novel MEK inhibitors, TAK-733 and CH4987655, from Takeda and Chugai/Roche respectively. The papers are informative in that they describe the somewhat different approaches taken by each of these groups. Both groups started from PD0325901, or close analogues thereof, and sought to remedy the reported metabolic instability of the hydroxamate group in these compounds. The Takeda group used the published x-ray co-crystal data (PDB1S9J) of one such analogue to design alternative templates in which the amide was incorporated into a fused bicyclic core, thus reducing the propensity for hydrolysis and the conformational flexibility of the molecule. The pyridopyrimidinedione system (see A below), incorporating the known hydrophobic, 2-fluoro-4-iodoaniline binding motif, appeared to provide a good fit with the allosteric site and the extra carbonyl group had the potential to form an additional H-bond to Val 211 and/or Ser212. Indeed, when prepared, this compound had very encouraging enzyme activity (IC50=26 nM) and provided the basis for a more detailed SAR exploration which in turn led to the discovery of TAK-733. The introduction of a polar sidechain allowed for additional interactions with Lys97 and the ATP phosphate, and presumably also modulation of physicochemical properties (although the paper doesn’t comment on this aspect), and an additional fluorine substituent also seems to have been beneficial with respect to potency. TAK-733 is highly potent (IC50=3.2 nM) and selective, has excellent PK in preclinical species, and has potent anticancer activity in various mouse xenograft models. The compound is suitable for once-daily oral dosing in humans and appears to be the subject of an ongoing trial.
The team at Chugai took a slightly different approach and chose to protect the hydroxamate group from hydrolysis by introducing substituents which would perturb the interactions with metabolising enzymes. They found that substitution (see PD0325901) in the 5-position was well tolerated and compounds of the general formula B had good potency and, as hoped, improved metabolic stability of the hydroxamate side-chain (relative to PD0325901 and analogues thereof). Further cycles of optimisation led to CH4987655 (see below), a potent and selective MEK inhibitor with good activity in xenograft models (particularly in combination with mTOR inhibitors), and a favourable PK profile in human volunteers. Interestingly, the authors also point out that CH4987655, in contrast to PD0325901, does not inhibit MEK in the brain (due to limited distribution from plasma) presumably with the intention of implying (although not explicitly stating) that this may have some bearing on some of the apparently centrally mediated side-effects noted with PD0325901. CH4987655 (redesignated as RO4987655) is also currently the subject of an ongoing clinical trial.