Rapid characterization of drug response using tumor-microenvironment-on-chip

Kyeonggon Shin, Purdue University


The use of microfluidic platforms has become an indispensable strategy to recapitulate physiological tumor microenvironment in order to describe cell and molecular biomechanics. To date, many promising early studies have characterized the role of the tumor microenvironment in cancer malignancy and investigated the efficacy of anti-cancer drugs. However, the essential in vivo conditions regulating tumor malignancy remains poorly understood. Additionally, the capability to predict the therapeutic efficacy of developed drugs is still limited. Consequently, a novel in vitro tumor model, entitled tumor-microenvironment on chip (TMOC), had been established to enable relevant in vivo tumor characterization within extracellular matrix (ECM) under physiological interstitial pressure and flow. In this study, TMOC was employed to investigate not only the role of regulation of tumor growth rate by 3D culture and interstitial flow but also the contribution of drug resistance to human tumors and drug binding, unbinding, and efflux rate constants in order to accelerate drug discovery. First of all, we found that growth rate comparison between monolayer (2D) and TMOC (3D) reflected that the role of extracellular matrix (ECM) was significant in tumor growth rate, however, the interstitial flow available on TMOC was not induce notable differences compared to tumors cultured without any flow. In the following study, exploring drug resistance in human cancer cells, we have observed that malignant MDA-MB-231 cells, triple negative and CD44 over-expressed human breast cancer cells were altered and became significantly resistant to doxorubicin(DOX) on TMOC compared with MCF-7 and SUM-159PT, whereas survival rate of MDA-MB-231 on 2D monolayer assays was significantly lower than MCF-7. Furthermore, in the case of DOX-loaded 250nm hyaluronic acid nanoparticles (HANP), which are designed to selectively bind to CD44 over-expressed cancer cells, the drug binding rate to MDA-MB-231 cells increased significantly compared to free DOX. Furthermore, a relatively higher drug unbinding and efflux rate was observed in DOX-HANP, accompanied by apparently higher survival rates compared to free DOX in spite of significantly more DOX-HANP accumulation in the cell area. This result infers that DOX-HANP is less able to facilitate drug binding to tumor nuclei. It may be because DOX-HANP was easily unbound from carrier protein due to its bigger size. Overall, these results suggest that the TMOC model is able to support the improved understanding of how tumors respond to anti-cancer drugs in vivo.




Han, Purdue University.

Subject Area

Biomedical engineering|Pharmacy sciences

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server