top of page


We employ a quantitative approach that integrates microfluidics, systems biology modeling, and in vivo experiments to investigate the role of complex microenvironments on cell growth, migration and response to perturbations in health (tissue homeostasis) and disease (cancer).

Microfluidic modeling of cell migration, cell-ECM and cell-cell interactions 

Microfluidics provides a powerful platform technology for developing a testbed to investigate fundamental mechanisms of cell-cell interactions in a complex microenvironment.

Our group has employed microfluidic devices to study the role of growth factor gradients, fluid flow, hypoxia, cell-cell signaling and 3D matrix mechanics on cell growth, migration and therapeutic responses.

We have shown how macrophages are recruited in response to paracrine factoors and interact with a 3D extracellular matrix in microfluidic devices (Fig. 1).


Figure 1: (A) 3D macrophage migration towards host cells in microfluidics. (B) Example images using RAW264.7 macrophages

Cellular dynamics in fibroblast-rich microenvironments

Advanced HER2+ breast cancer has a poor prognosis and improved patient outcomes will depend on a better understanding of mechanisms of therapy resistance. Stromal fibroblasts, an abundant cell in the tumor microenvironment, have been linked with poor treatment response in HER2+ breast cancer patients.


Motivated by our in vitro studies on tumor-fibroblast interactions, we hypothesized that factors in the breast cancer stroma activate pro-survival signaling and contribute to therapy resistance (Fig 2)To identify cellular programs that confer resistance to targeted therapies and design effective treatments to restore sensitivity we will:

  • Develop novel microfluidic platforms to culture tumor explants

  • Measure tumor-fibroblast dynamics and develop mathematical models (agent-based) of their interactions

  • Systems analysis of tumor-fibroblast signaling

This integrated computational-experimental approach can address a major challenge in drug resistance: dissecting the effects of tumor cell intrinsic (e.g. genetic background) versus extrinsic factors (e.g. stroma-secreted factors) on drug sensitivity. In situ analyses  will provide important insights into stromal factors that activate tumor cell signaling pathways, and will also serve as a comprehensive dataset for follow-up studies to dissect paracrine vs. juxtacrine signaling and develop computational models of the tumor microenvironment.


Taken together, these studies will identify biomarkers to guide therapy and will also lay the foundation for further investigations of factors conferring drug resistance (e.g. distinct microenvironments, role of additional cell types).  Funding: K99/R00 award from NCI (K99CA222554)


Figure 2: (A) New microfluidic culture platforms to study cellular dynamics in tumor explants. (B) Agent-based modeling
(C) Cyclic immunofluorescence for multiplex protein expression analysis 

Video 1: Breast cancer cells (green) interacting with fibroblasts (white) under baseline conditions

Modeling of heterotypic cancer-stromal cell crosstalk and biophysical forces in ovarian cancer


Figure 3: (A-B) 3D microfluidic model of ovarian cancer metastasis. (B) Histology analysis of ovarian cancer invasion in patient-derived xenografts (PDX). 

Ovarian cancer is oftentimes not detected until after metastases have occurred. The mechanisms of tumor cell survival during metastatic dissemination remain poorly understood, and while both biomechanical forces and stromal factors in the microenvironment have been implicated, further research is needed to understand their role. A better understanding of these processes can uncover novel treatment targets to block invasion.


In order to address this need, our lab has developed a 3D microfluidic platform to model the metastatic cascade and control for different microenvironmental factors (Fig 2). We have also characterized the growth, protein expression and drug sensitivity of a panel of 14 high-grade serous ovarian cancer patient-derived xenografts (Zervantonakis et al 2017, Liu and Palakurthi 2016). Our specific aims for this project include: 

  • Role of macrophages and fluid flow-generated forces on ovarian cancer cell adhesion and invasion

  • Strategies to restore therapeutic sensitivity on ovarian cancer implants

These studies will identify critical factors in the metastatic microenvironment that regulate cancer invasion and uncover biomarkers to predict metastatic potential. 

Novel localized drug release technologies in microfluidics:

We have developed a novel acoustofluidic platform (Zervantonakis et al 2016 : coupling of microfluidics with focused ultrasound) that allows for studying the transport of heat-sensitive nanomedicines (e.g. Thermodox) and monitoring in real-time cellular behavior (Fig 4).



Figure 4: (A) Localized drug release in tissues  (B) Acoustofluidic platform
(C) Results from in vitro assays using a glioblastoma cell line

Single-cell assay development

Intra-tumor heterogeneity represents a major challenge in cancer treatment and is linked to the development of resistance. Understanding how different tumor cells respond to treatment can lead to the development of new combination therapies to specifically target resistant subpopulations. 

We have developed PDMS-based microwell systems (see video on the right) that allow for monitoring growth of ovarian cancer cells in suspension and interactions with macrophages (see video below). 

Video of macrophages (GFP) interacting with ovarian cancer cells under suspension culture conditions

We have employed this single-cell culture platform to study ovarian cancer cell survival in suspension (Iwanicki, Chen, Iavarone, Zervantonakis). We found that mutant p53 enables survival/spheroid formation of single cancer cells in suspension, suggesting that early ovarian cancer dissemination may arise from a single cells.

Ongoing studies in the group including co-culture assays of ovarian cancer cells with macrophages (see video on the left) and analysis of heterogeneous spheroid populations from ovarian cancer patient-derived models.

bottom of page