Ponente
Descripción
Melanoma, among different types of cancer, distinguishes for its high aggressiveness and strong metastatic potential. During the last few years, mathematical modeling methods have been widely used to describe tumor growth dynamics and to analyze the effectiveness of different therapeutic approaches [1]. the objectives of this study is to use computational methods for the calibration and selection of mathematical models based on ordinary differential equations (ODEs), utilizing experimetal volume data from [2], in order to evaluate the predictive capacity of the selected model and to perform computational simulations under different therapeutic conditions.
In the first stage of the study, mathematical models of tumor growth were used, (Exponential, Logistic, Von Bertalanffy, Linear, Mendelssohn, Surface, and Gompertz models), to describe melanoma progression since experimental data originating from murine models with untreated tumors [2]. The parameters of each model were estimated using nonlinear fitting techniques based on the Least Squares Method. The numerical solution of the differential equations was performed using the LSODA solver, an adaptive method capable of automatically alternate schemes suitable for rigid and non-rigid systems. The selection of the most appropriate model was conducted based on statistical criteria, as the BIC (Bayesian Information Criterion) and the AIC (Akaike Information Criterion), which balances adjustment quality and model complexity, in addition to other metrics. For this stage, specific data from the NOG control group were used, consisting of immunodeficient mice lacking interleukin-2 (IL-2) expression [2].
Therefore, the selected growth function will serve as the basis for constructing the complete model, in which an additional term representing the effect of immunotherapy will be incorporated. For this purpose, data from the NOG CAR-T group will be used consisting of immunodeficient mice lacking interleukin-2 (IL-2) expression, submitted to immunotherapy with CAR-T cells (genetically modified T lymphocytes to express chimeric antigen receptors capable of recognizing and eliminating specifically tumor cells) [2].
The study establishes a relationship between the two developed stages, in which the modeling of untreated tumor growth underlies the analysis of the therapeutic response. In future work, it is intended to relate all the stages of the model more integrated with the objective to improve its predictive capability under different therapeutic conditions. These extensions may contribute to a more comprehensive understanding of the interaction between the tumor and immunotherapy. Moreover, the proposed approach enables the quantitative evaluation of deviations related to the natural tumor growth induced by therapy, providing a tool to investigate the impact of melanoma dynamics treatment.
References:
[1] ALVES, N. B.; MANCERA, P. F. A.; RODRIGUES, G.; ANTUNES, M. E. Model selection of tumor growth dynamics in untreated melanoma as a base linear for CAR-T cells therapy. Hematology, Transfusion and Cell Therapy [S.l.], v. 48, supl.1, p.106272, 2026. DOI:https://doi.org/10.1016/j.htct.2026.106273.
[2] FORSBERG, E. M. V.; LINDBERG, M. F.; JESPERSEN, H.; ALSÉN, S.; BAGGE, R. O.; DONIA, M.; SVANE, I. M.; NILSSON, O.;NY,L.;NILSSON,L.M.;NILSSON,J. A. HER2 CAR-T cells eradicate uveal melanoma and T-cell therapy–resistant human melanoma in IL2 transgenic NOD/SCID IL2 receptor knockout mice. Cancer Research, [S.l.], v. 79, n.4, p.899–904, 2019. DOI: https://doi.org/10.1158/0008-5472.CAN-18-3158.