AbstractIn metastatic prostate cancer, most cases of resistance to direct AR inhibitors (ARi) are associated with a rebound in serum PSA implying AR reactivation. Resistance to first generation ARi, like bicalutamide, has been linked to AR amplification and is thought to be caused by increased levels of AR protein favoring the drug’s partial agonist activity. This agonism counteracts inhibition with a net effect of incomplete antagonism. The phenomenon of anti-androgen withdrawal syndrome documented with early ARi supports this hypothesis and presumably reflects net agonist activity. Enzalutamide ushered in the current generation of approved ARi and was identified using prostate cancer cell line models of bicalutamide resistance with increased AR protein. It has recently become clear that a large majority of cases of mCRPC resistance to enzalutamide is associated with even higher copy number amplification of AR. Enzalutamide, and other latest generation ARi, including AR degraders in clinical development, all share a closely related (‘lutamide) core chemical structure with the earlier generation partial agonists implying a similar mechanism of binding AR. This raises the possibility that high copy AR amplification-driven resistance to enzalutamide is due to an antagonism-to-agonism switch mechanistically identical to that of bicalutamide but requiring higher AR levels. However, AR levels exceeding those required to confer bicalutamide resistance are detrimental to standard cancer cell lines growing in culture. Here we describe new mechanistic models of androgen-stimulated AR activation across a wide range of AR levels. In these models early generation ‘lutamides, but not enzalutamide, show a switch to agonist like behavior at moderate levels of AR overexpression. At higher levels of AR, enzalutamide, apalutamide, and darolutamide show the same agonist conversion and corresponding evidence of incomplete antagonism in enzalutamide-resistant prostate cancer cells with high copy AR amplification. We employed these assays in concert with a novel, in-cell, chemoproteomic screening approach to identify non-lutamide ARi capable of complete antagonism. We discovered a novel chemotype of ARi that lacks any propensity for an agonism switch, even at high levels of AR protein. These compounds completely suppress viability of VCAP prostate cancer cells with high copy AR amplification, have no effect on the viability of AR-independent cell lines, and show activity against all clinically common AR point mutants. These observations suggest that the potential for a switch to agonism is inherent, in the ‘lutamide scaffold, and is exacerbated by increased AR levels leading to incomplete antagonism and resistance. This resistance can be effectively combated with a new class of complete AR antagonists.Citation Format:Anthony Szempruch, Trevor Scott, Justin Ernst, Jessica Hoffman, Amy Klova, Amy Liu, Nhin Lu, Angelica Milik, Sherry Niessen, Garrick Packard, Jarrett Remsberg, Roksolana Sipos, Tiffany Soriano, Gary O’Neill, Sean Buchanan. Discovery of a novel class of androgen-competitive AR inhibitors that can combat drug resistance due to AR gene amplification [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 4295.