Background: Chemosensitization by azacitidine (AZA) priming has been demonstrated preclinically and is now supported by clinical data, but the underlying mechanism for the priming effect is not well understood. Previous in vitro studies of priming have focused on individual cell lines and specific epigenetically-silenced genes; however, we show that sensitization is not a universal effect of AZA pretreatment in all cancer cell lines or with all combination agents. Studying the molecular differences between cell lines that are sensitized and those that are not may yield insights into the mechanisms of priming.

Purpose: To identify solid cancer cell lines that consistently show AZA sensitization to platin-induced cytotoxicity, for future molecular characterization of the priming mechanism.

Methods: Ninety-two solid cancer cell lines, from seven indications, were treated with vehicle or 1μM AZA for 24h before being treated with carboplatin (CARB), cisplatin (CIS), or Abraxane (ABX) for 72h. CellTiter-Glo was used to measure cell viability. The fractional product method and normalized dose-response curves were used to characterize the drug combination interaction. In subsequent experiments, various AZA pretreatment durations (4-72h) were tested. DNA was prepared for DNA methylation (LINE-1) analysis, and cell lysates were harvested for DNMT1 Western blotting.

Results: In a subset of cancer cell lines, AZA pretreatment resulted in enhanced sensitivity of cells to CARB and CIS. Seven cell lines across 5 cancer indications consistently showed a sensitization effect with AZA priming, decreasing the CARB IC50 by a mean of 59.6% (range 52.9-68.4%) and CIS IC50 by a mean of 44.6% (range 36.2-54.6%). Priming for CARB was concordant with priming for CIS, but not for ABX, indicating that the AZA pretreatment does not globally increase the susceptibility of cells to cytotoxic agents through non-specific toxic effects; rather, these results suggest that priming by AZA has specific pharmacodynamic effects unique to the combination agent. Follow-on experiments demonstrated that AZA pretreatment of cells for 18-24h was required for priming, whereas the magnitude of priming was not increased with longer (48-72h) AZA pretreatment durations. Effects of AZA on DNMT1 depletion and DNA hypomethylation were detected in all cell lines tested. The magnitude of these proximal PD effects did not correlate with priming, suggesting that differences in sensitization cannot be attributed to differential drug uptake and DNA incorporation.

Conclusions: We identified a panel of solid cancer cell lines that consistently demonstrate sensitization to CARB and CIS with AZA priming. By comparing genetic and epigenetic profiles between these lines and a closely matched set of cell lines that does not show priming, we hope to generate clinically testable hypotheses for predictors of sensitization in patients.

Citation Format: Aaron N. Nguyen, Manith Norng, Antonio Luna-Moran, Kyle J. MacBeth, Jorge F. DiMartino. Development of a robust preclinical model for studying the mechanism of azacitidine priming for platin-induced cytotoxicity. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 4656. doi:10.1158/1538-7445.AM2013-4656