Combined analysis of tumor size data and histological biomarkers drives the development of a semi-mechanistic model of the effect of the antiangiogenic drug Sunitinib in mice
Ribba B,, Watkin E., Tod M., Girard P., Grenier E., You B., Wei M., Giraudo E., Freyer G.
INRIA Rhône-Alpes, NUMED project team, Ecole Normale Supérieure de Lyon, 46 allée d’Italie, 69007 Lyon Cedex, France.
Objectives: One difficulty in antiangiogenic drug development is the proper evaluation of drug efficacy. In preclinical research, biomarkers from histological analysis are extensively used, however, the information they provide is incomplete. Preclinical development of antiangiogenic drug may beneficiate from drug-disease modeling approaches incorporating histological data indicative of cytostatic action. In this study, we developed a semi-mechanistic model of the effect of the antiangiogenic drug Sunitinib in mice combining tumor size data and histological biomarkers.
Methods: Human colorectal cancer cells HT-29 were implanted in 45 athymic mice. When tumor volume reached 200-300 mm3, mice were randomized in two treatment groups. In the first group (n=15), mice received a single oral dose of 40 mg/kg of Sunitinib. In the second group (n=30), mice received the same dose daily, for twelve days. Tumor diameters were periodically recorded for each animal. Mice were randomly euthanized at different time points following the drug administration and tumors were analyzed: Intratumoral blood vessel density and diameter as well as the percentage of hypoxic and necrotic tissue were assessed. Monolix 2.4 was used to estimate the parameters of the mixed-effect models. Model selection was based on standard errors of the estimates, goodness of fit plots, visual predictive check as well as shrinkages.
Results: Tumor size data were first analyzed alone (single-output model). We adapted a model formerly developed by Hahnfeldt and colleagues . In this model, the limiting size for the tumor is controlled by a time-dependent variable, namely K, or carrying capacity, which increases as tumor grows to account for the process of angiogenesis. Sunitinib action was modeled using a kinetic-pharmacodynamic (K-PD) formulation  and assuming the antiangiogenic drug to induce a decline in the tumor carrying capacity K, and consequently tumor growth inhibition and eventually shrinkage. We sequentially extended this model to account for the histological data. In this semi-mechanistic (four-output) model, necrotic tissue forms as a result of hypoxia as the tumor reaches its limiting size K.
Conclusions: The main innovation of this semi-mechanistic model lies in its ability to integrate classical histological biomarkers such as those commonly retrieved in preclinical studies. The model provides insights into the mechanisms of tumor growth inhibition and shrinkage following antiangiogenic drug administration and in this respect, may help the development of such compounds in preclinical stages.
 Hahnfeldt P., Panigraphy D., Folkman J., Hlatky L. Tumor development under angiogenic signaling: a dynamical theory of tumor growth, treatment response, and postvascular dormancy. Cancer Res 1999; 59(19):4770-5.
 Jacqmin P., Snoeck E., van Schaick E.A., Gieschke R., Pillai P., Steimer J.-L.; Girard P. Modelling response time profiles in the absence of drug concentrations: definition and performance evaluation of the K-PD model. J Phamacokinet Phamacodyn 2007; 34(1):57-85.