Is the Fate of Clinical Candidate Arry-520 Already Sealed? Predicting Resistance in Eg5-Inhibitor Complexes.

Abstract

Arry-520 is an advanced drug candidate from the Eg5 inhibitor class undergoing clinical evaluation in patients with relapsed or refractory multiple myeloma. Here, we show by structural analysis that Arry-520 binds stoichiometrically to the motor domain of Eg5 in the conventional allosteric loop L5 pocket in a complex that suggests the same structural mechanism as other Eg5 inhibitors. We have previously shown that acquired resistance through mutations in the allosteric-binding site located at loop L5 in the Eg5 structure appears to be independent of the inhibitors' scaffold, which suggests that Arry-520 will ultimately have the same fate. When Arry-520 was assessed in two cell lines selected for the expression of either Eg5(D130A) or Eg5(L214A) STLC-resistant alleles, mutations previously shown to convey resistance to this class of inhibitors, it was inactive in both. Surprisingly, when the cells were challenged with ispinesib, another Eg5 inhibitor, the Eg5(D130A) cells were resistant, but those expressing Eg5(L214A) were strikingly sensitive. Molecular dynamics simulations suggest that subtle differences in ligand binding and flexibility in both compound and protein may alter allosteric transmission from the loop L5 site that do not necessarily result in reduced inhibitory activity in mutated Eg5 structures. Although we predict that cells challenged with Arry-520 in the clinical setting are likely to acquire resistance through point mutations in the Eg5-binding site, the data for ispinesib suggest that this resistance mechanism is not scaffold independent as previously thought, and new inhibitors can be designed that retain inhibitory activity in these resistant cells.