Abstract 3980: Computational analysis of paclitaxel-induced tumor priming that leads to enhanced deposition and efficacy nanoparticulate doxorubicin

Abstract

Abstract Purpose: Paclitaxel (PAC) has been shown to cause tumor priming (TP), which increases delivery of subsequently-administered drugs. A wave of PAC-mediated apoptosis results in a transient reduction in tumor cell density, expansion of the interstitial space, and vascular permeability compromise that enhances tumor deposition of nanoparticulate formulations such as liposomal doxorubicin (L-dox). Our aim was to develop an integrated quantitative pharmacokinetic/pharmacodynamic (PK/PD) analysis approach to optimize combination chemotherapy of PAC with L-dox. Methods: Data were extracted from the study of Lu et al. (J Pharmacol Exp Ther 322: 80, 2007) in which mice bearing human tumor xenografts were treated with PAC (40 mg/kg), L-dox (20 mg/kg), TP dosing (PAC followed by L-dox at 48h, the peak of the priming effect), or the reverse sequence (L-dox then PAC). Data included plasma and tumor drug concentrations, kinetics of PAC and L-dox induced apoptosis, and % change in tumor volume from baseline. Drug profiles were described with a PK model for carrier-mediated PAC, and a two-compartment model for L-dox, with first-order release of free drug from the liposome. Intratumor concentrations were described using a hybrid well-stirred model with estimated partition coefficients (kp) for each drug. Tumor PK profiles were used to drive apoptotic responses, which were modeled using nonlinear time dependent signal transduction functions. The tumor growth PD model consisted of two transit compartments, with a first order net growth rate constant and the apoptotic signals from each agent driving cytotoxic effects. PAC induced TP was modeled using a feedback loop, with the apoptosis signal of PAC enhancing the deposition of L-dox (i.e. increasing kp for dox). Model fittings were performed using MATLAB, and simulations were conducted to explore priming sequences. Results: The final model captured well the PK of PAC and L-dox in plasma and tumors, and the time course of apoptosis induction, and tumor growth for each treatment sequence. With single agent dosing, kp values for PAC and L-dox were estimated at 0.044 and 0.085. L-dox kp increased 2-fold after the TP treatment. Apoptosis signals exhibited a delayed onset that was well captured, and the intratumor concentrations producing maximum effects (Emax) and 50% Emax were 18 and 7.2 μg/mL (PAC) and 17.6 and 14.3 μg/mL (L-dox). The duration of drug induced apoptosis was 27.4 h for PAC and 15.8 h for L-dox. Simulation with the PK/PD model suggested that earlier administration of L-dox would increase efficacy markedly. Conclusions: A model was successfully developed that captured the priming effect on PK and efficacy. Simulation suggested that administration of L-dox 24 h prior to the priming peak would enhance efficacy further. This model could be adapted for evaluating other combination chemotherapies using PAC as a TP agent. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 3980. doi:1538-7445.AM2012-3980