(T-093) Preclinical tumor growth inhibition modeling and simulation to support dosing regimen selection of a novel EWS-FLI1 inhibitor for Ewing sarcoma treatment

Markos Leggas, Ph.D. – Member, Department of Pharmacy and Pharmaceutical Sciences, St. Jude Children's Research Hospital

A novel EWS-FLI1 inhibitor, an analog of mithramycin, for Ewing sarcoma (ES) treatment was studied in a murine xenograft model of ES to evaluate its anti-tumor efficacy. The objective of this analysis was to develop a tumor growth inhibition (TGI) model to characterize the pharmacokinetic-pharmacodynamic (PK/PD) relationship of this analogue and apply the TGI model to simulate tumor regression at different doses, schedules, and cycles to propose an appropriate dosing regimen for future efficacy studies.

Tumor growth data were collected from anti-tumor efficacy studies conducted in TC-32 ES tumor-bearing athymic nude mice. Mice were treated with vehicle or 0.3, 0.6, and 0.9 mg/kg of the mithramycin analog on a daily IV bolus schedule for five days (QDx5). PK was studied separately at three doses (0.3, 1, and 3 mg/kg), and parameters were estimated using population PK modeling. A sequential modeling approach was adopted, starting with PK modeling followed by TGI modeling with fixed PK parameters. The Simeoni tumor growth model was used to describe natural tumor growth1. The drug effect causing tumor growth inhibition was integrated into the model as a linear effect dependent on concentration (Equation 1). The predictive capability of the model was assessed by comparing simulated tumor growth inhibition of a 2.4 mg/kg Q3Dx6 dosing schedule with experimental data, which was not included in model development. To propose an optimal dosing regimen for future efficacy studies, various tolerated dosing regimens were explored through TGI model-based simulation, followed by survival analysis of the simulated data, to identify dosing regimen(s) that yield prolonged tumor regression and improved survival.

dTV/dt=(kge×TV(t))/〖(1+〖(kge/kgl×TV(t))〗²⁰)〗^1/20 -kkill×C(t) ×TV(t) (Equation 1)

A two compartment PK model adequately described the PK of this analogue with clearance (CL) 40.9 ml.h-1kg-1 (RSE=7.48%), central volume of distribution (V1) 25.36 ml.kg-1 (RSE=8.55%), intercompartmental clearance (Q) 0.76 ml.h-1kg-1 (RSE=23.3%) and peripheral volume of distribution (V2) 6.26 ml.kg-1 (RSE=18.6%). The TGI model adequately characterized the natural tumor growth over time with an exponential growth rate (kge) 0.008 h-1 (RSE=6.64%) and a linear growth rate (kgl) 6.18 mm3.h-1 (RSE=13.3%) as well as the drug effect (kkill) 0.000028 ml.ng-1.h-1 (RSE=9.31%). The observed tumor growth inhibition in mice treated with 2.4 mg/kg Q3DX6 was well predicted by the TGI model, suggesting the predictive utility of this model. TGI model-based simulations for different dosing regimens showed 2.4 mg/kg at Q3Dx5 schedule with two 21 day cycles would lead to the longest tumor regression and highest median survival ( >100 days) suggesting that this analogue has a good therapeutic window.

This TGI model based prediction will be tested experimentally.

Citations: [1] Simeoni M, et al. Cancer Research, 2004; 64(3):1094-1101