Quantitative investigation of ATP switch concept of STA551 by combined in vivo tissue distribution study and physiologically-based pharmacokinetic modeling

Objectives: STA551 is an anti-human CD137 (hCD137) switch antibody with ATP-dependent antigen binding that intends potent anti-tumor effects and superior safety [ 1]. This novel switch antibody can also bind to hCD137 in the presence of ADP and AMP as well as ATP (ATP switch molecules). Sta-MB, mouse surrogate antibody of STA551, has shown strong anti-tumor effects in several tumor-bearing mouse models, suggesting high concentration of ATP switch molecules in the tumor interstitial space. However, quantitative measurement of ATP switch molecules is highly challenging as these molecules can be easily hydrolyzed. This study aims to quantitatively investigate ATP switch molecule concentration in the tumor interstitial space and demonstrate the ATP switch concept of STA551. Methods: ATP switch molecule concentration in the tissue interstitial space was investigated by combining tissue distribution study and physiologically-based pharmacokinetic (PBPK) modeling. Iodine-labeled conventional anti-hCD137 antibody, 125I-Ure-mIgG1, was intravenously administered as a tracer to all tumor-bearing hCD137 knock-in mice. In some mice, unlabeled Ure-mIgG1 at 20 mg/kg or Sta-MB at 1 or 20 mg/kg was co-administered. The tracer radioactivity levels in plasma, blood and tissues were monitored. A PBPK model was established using MATLAB/Simbiology in order to describe the obtained distribution data. The established PBPK model has major normal tissues and tumor with tissue intrinsic clearance and target-mediated clearance. ATP-dependent antigen binding by Sta-MB was expressed by Emax equation. PBPK model parameters including tissue intrinsic clearance, interstitial distribution clearance, target related parameters and ATP switch molecule concentration were estimated to explain the distribution data. Results: Ure-mIgG1 co-administration at 20 mg/kg increased plasma tracer radioactivity levels and decreased the tissue-to-plasma (T/P) ratio in several normal tissues and tumor, suggesting antigen expression in these tissues. However, Sta-MB co-administration at 1 or 20 mg/kg showed little effect on plasma radioactivity levels but clearly decreased T/P ratios, mainly in tumors, strongly suggesting tumor-selective binding of Sta-MB. PBPK model parameters including ATP switch molecule concentration were successfully estimated. The estimated ATP switch molecule concentration in tumors was much higher than in non-tumor tissues. Conclusions: This integrated approach combining tissue distribution study and PBPK modeling successfully estimated ATP switch molecule concentration in the tumor interstitial space and demonstrated the ATP switch concept for STA551 in vivo. Furthermore, these findings are expected to be utilized as information for the early clinical development plan for STA551.