Predictive Computational Modeling: In Vivo response with In Vitro results in Breast Cancer

Nearly 12% of women in the United States will develop breast cancer during their lifetime. Therefore, the significance of treating breast cancer with both currently available and developing treatment methods is undisputable. Unfortunately, treatment results tested in laboratories that are often expected to be similar to results in patients often do not correlate with the latter results. Although it is easier, less expensive, and more convenient to perform in-vitro, or laboratory, experimentation, it is necessary to convert these results into plausible results in a three-dimensional in-vivo patient environment. Animal models can offer insight but would be difficult to perform for each patient. Therefore, the purpose of this project was to predict tumor growth in-vivo through in-vitro treatment results of the gemcitabine drug using a mathematical model. The hypothesis was that although the in-vitro and in-vivo results would be somewhat correlated in terms of tumor growth, the rate of tumor growth in-vivo would eventually begin to decrease due to insufficient nutrient uptake from tumor vasculature. Histological samples of breast cancer stained with the CD31 immunostain were analyzed to supplement the previously mentioned treatment results; these results were used to determine values for the variables of the mathematical model, which was implemented with the MATLAB computing language. The results supported the hypothesis: as the tumor radius increased, the rate of tumor growth was predicted by the model to decrease to a point of negative growth except in the case of very low cellular apoptosis. This procedure, therefore, may be able to help bridge the gap between expected in-vitro breast cancer treatment results and actual in-vivo results.