BackgroundENZA and APA are androgen receptor inhibitors approved for treatment of castration-resistant prostate cancer (CRPC) and nonmetastatic CRPC, respectively. Their chemical structures differ with ENZA having 2-cyanophenyl and dimethyl moieties and APA 2-cyanopyridyl and cyclobutyl moieties. In the SPARTAN clinical trial, APA treatment was reported to be associated with an increased incidence of skin rash compared with the placebo group, which was not observed with ENZA in the PROSPER trial in a similar patient population; therefore, we examined their in vitro chemical reactivity and hypersensitivity potential in an MDAM. MethodsTo assess possible reactivity of cyanopyridine, in vitro studies examining the chemical stability in buffer (+/- glutathione) and covalent binding to proteins (eg, bovine serum albumin [BSA]) were evaluated. For the MDAM, C57BL/6 mice were injected subcutaneously with APA, ENZA, or RD162 (Analogue 1), daily for 3 days at 25, 50, or 100mg/kg/day. Brachial lymph nodes were harvested on day 6 and lymph node cellularity was analyzed by flow cytometry. ResultsThe 2-cyanopyridine moiety of APA was found to be chemically reactive with the thiol nucleophile glutathione, resulting in rearranged thiazoline products. Radiolabeled APA, but not ENZA, was shown to react with mouse and human plasma proteins. Thiol nucleophiles decreased the extent of covalent binding to the model protein BSA while amine and alcohol nucleophiles had no effect, suggesting a reaction with cysteine sulfhydryl groups of proteins. In the MDAM, dose-dependent higher lymph node cellularity was observed for the APA groups (compared with the vehicle control), suggesting immune activation. ENZA and Analogue 1 demonstrated substantially less covalent binding activity and no change in lymph node cellularity (compared with vehicle) in the MDAM. ConclusionsThese data support the hypothesis that the 2-cyanopyridine moiety present in APA, but absent in ENZA or Analogue 1, may react with cysteine residues in proteins to form haptens that may trigger an immune response leading to the increased incidence of skin rash clinically observed in patients treated with APA, but not with ENZA. Editorial acknowledgementMedical writing and editorial assistance funded by Pfizer Inc. was provided by Ira Mills, PhD, and Michele Salernitano from Ashfield Healthcare Communications. Legal entity responsible for the studyPfizer Inc. and Astellas Pharma, Inc. FundingPfizer Inc. and Astellas Pharma, Inc. DisclosureM. Guha: Shareholder / Stockholder / Stock options, Employee: Pfizer Inc.. C. Ji: Shareholder / Stockholder / Stock options, Employee: Pfizer Inc.. X. Zhu: Shareholder / Stockholder / Stock options, Employee: Pfizer Inc.. J. Whritenour: Shareholder / Stockholder / Stock options, Employee: Pfizer Inc.. M. Hemkens: Shareholder / Stockholder / Stock options, Employee: Pfizer Inc.. S. Tse: Shareholder / Stockholder / Stock options, Employee: Pfizer Inc.. G.S. Walker: Shareholder / Stockholder / Stock options, Employee: Pfizer Inc.. E. Evans: Shareholder / Stockholder / Stock options, Employee: Pfizer Inc.. N.K. Khan: Shareholder / Stockholder / Stock options, Employee: Pfizer Inc.. M.B. Finkelstein: Shareholder / Stockholder / Stock options, Employee: Pfizer Inc.. E. Callegari: Shareholder / Stockholder / Stock options, Employee: Pfizer Inc.. R..S. Obach: Shareholder / Stockholder / Stock options, Employee: Pfizer Inc..