Tagliavini A. (1) , Borella E. (1), Piana C. (1), Sanna M.D. (1), Mazzei P. (1), Troconiz I.F. (2), Windak R.(3), Brzózka K. (3), Baldini S. (1), Goso C. (1), Merlino G. (1), Tomirotti A. (1), Tagliacozzi D. (1), Capriati A. (1), Pellacani A. (1)
(1) Menarini Ricerche SpA, Italy, (2) Pharmacometrics & Systems Pharmacology, Department of Pharmaceutical Technology and Chemistry, School of Pharmacy and Nutrition, University of Navarra, Spain, (3) Selvita S.A., Poland
Objectives:
In the preclinical phase of oncology drug development, a common way to test the activity of a drug is by means of tumor xenograft experiments. Of particular interest is the possibility of integrating the drug-to-biomarker dynamics with the associated drug-to-tumor growth inhibition response. This would allow to predict tumor response from biomarker data, which would be a key in the translation from preclinical to clinical research. The aim of this analysis is to find a quantitative relationship between MEN1703 drug exposure or plasma/tumor partitioning coefficient and pharmacological effects as measured by biomarkers and tumor growth inhibition in MOLM16 leukemia (AML) cell line xenografts in mice. To address this aim, a predictive pharmacokinetic/pharmacodynamic (PK/PD) model, which integrates preclinical pharmacokinetic, biomarker and efficacy data has been developed.
Methods:
First, MEN1703 Pharmacokinetics (PK) model in mouse was developed using data both at single and multiple doses from four studies. Second, PK data in plasma and tumor from two studies were used to determine the relationship between MEN1703 concentrations in the two matrices, with biomarker data in tumor. Third, a model describing dynamics of S6 (Ser235/236) phosphorylation inhibition (%) in MOLM-16 xenograft was developed by using parameter estimates from the plasma-tumor PK model. Fourth, xenograft tumor growth and MEN1703 tumor growth inhibition data from four studies in mouse were modelled by means of the modified biomarker-driven TGI model developed by Simeoni et al.[1] and Sardu et al. [2]. All the analyses were performed using the NONMEM 7.3 computer program. Selection between models was based mainly on the goodness of fit and residual plots, and precision of parameter estimates expressed as coefficient of variation
Results:
Disposition of MEN1703 in mouse plasma was best described with a one compartment model with a linear elimination. Predicted plasma concentrations of MEN1703 were related with concentrations in tumor through a partition coefficient between plasma and tumour (kp ~10). The time course of biomarker pS6 (Ser235/236) inhibition was best described with a direct response model driven by MEN1703 concentrations in tumour. The IC50 and γ estimated values were 7360 ng/mL and 3.5, respectively. Data obtained from different TGI studies were fit separately due to high inter-study variability. The tumour growth in the absence of MEN1703 administration was well described by the model allowing for the switch from an exponential to a linear growth. With regards to the perturbed growth model, the Simeoni model captured well the tumor growth profiles and the effect of the anticancer treatment k2 for all the studies which ranged from 0.016 to0.043 μg-1·mL·d-1. Standard goodness of fit, together with prediction-corrected VPC were generated for each study. Overall, the models were able to well predict the observed tumor growth data.
Conclusions:
An integrated PK-biomarker-efficacy model for MEN1703 has been developed in mouse. This preclinical model-based framework in conjunction with a population PK model in human will be used to support dose selection in clinical trials.
References:
[1] Simeoni M, Magni P, Cammia C, De Nicolao G, Croci V, Pesenti E, Germani M, Poggesi I, Rocchetti M. Predictive pharmacokinetic-pharmacodynamic modeling of tumor growth kinetics in xenograft models after administration of anticancer agents. Cancer research. 2004 Feb 1;64(3):1094-101.
[2] Sardu M.L., Poggesi I., De Nicolao G. Biomarker- versus drug-driven tumor growth inhibition models: an equivalence analysis. J Pharmacokinet Pharmacodyn 42:611–626, 2015
Reference: PAGE 28 (2019) Abstr 9127 [www.page-meeting.org/?abstract=9127]
Poster: Drug/Disease Modelling - Oncology