Optimal Therapy Design Using Artificial Neural Network Surrogate Models of Fluid and Solute Transport in Tumors | AIChE

Optimal Therapy Design Using Artificial Neural Network Surrogate Models of Fluid and Solute Transport in Tumors

The application of quantitative and formal methods in cancer research can enable a fundamental understanding of the limitations of conventional therapies, uncover novel treatment strategies, and help optimize therapy for individual patient outcomes. Particularly, normalization of the tumor microenvironment (TME) has been proposed to improve the delivery of oxygen, antibodies, and nanomedicines to solid tumors and in turn improve treatment efficacy. A physiological transport framework based on a mechanistic model has been established to simulate the fluid and macromolecule transport in a tumor (Baxter et al). This model was used to solve parameter estimation problems to fit the model against in vivo experimental data using deterministic global optimization. However, it is found that the parameter estimation problems are computationally expensive to achieve global optimality using currently developed bounding routines for the mechanistic model. In this work, we propose to use artificial neural networks (ANN) to establish surrogate models for solving the parameter estimation problems to global optimality. In addition, inequality constraints are added on the feasible set of these problems based on the interstitial fluid pressure data (Martin et al). Furthermore, we also propose to use the ANN surrogates to enable the solution of a simultaneous dosage and macromolecule design problem for better patient outcomes. The use of ANN models has been confirmed to provide accurate solutions for solving challenging global optimization problems with substantially accelerated speed. The experimentally validated computational modeling approach enables TME-normalizing therapy design and anticancer drug design as well as provides insight into how in silico modeling approaches aid in predicting dose response in preclinical studies.

References:

Baxter LT, Jain RK. Transport of fluid and macromolecules in tumors. I. Role of interstitial pressure and convection. Microvascular research 1989;37(1):77–104.

Baxter LT, Jain RK. Transport of fluid and macromolecules in tumors. II. Role of heterogeneous perfusion and lymphatics. Microvascular research 1990;40(2):246–263.

Baxter LT, Jain RK. Transport of fluid and macromolecules in tumors: III. Role of binding and metabolism. Microvascular research 1991;41(1):5–23.

Baxter LT, Jain RK. Transport of fluid and macromolecules in tumors. IV. A microscopic model of the perivascular distribution. Microvascular research 1991;41(2):252–272.

Martin JD, Panagi M, Wang C, Khan TT, Martin MR, Voutouri C, et al. Dexamethasone Increases Cisplatin-Loaded Nanocarrier Delivery and Efficacy in Metastatic Breast Cancer by Normalizing the Tumor Microenvironment. ACS Nano 2019 jun;13(6):6396–6408.