Tumor Microenvironment on a Chip: The Progress and Future Perspective

Abstract

Tumors develop in intricate microenvironments required for their sustained growth, invasion, and metastasis. The tumor microenvironment plays a critical role in the malignant or drug resistant nature of tumors, becoming a promising therapeutic target. Microengineered physiological systems capable of mimicking tumor environments are one emerging platform that allows for quantitative and reproducible characterization of tumor responses with pathophysiological relevance. This review highlights the recent advancements of engineered tumor microenvironment systems that enable the unprecedented mechanistic examination of cancer progression and metastasis. We discuss the progress and future perspective of these microengineered biomimetic approaches for anticancer drug prescreening applications.

Keywords: drug screening; in vitro disease models; microfluidics; nanomedicine; organ-on-a-chip; tumor microenvironment.

Figure 1

The tumor microenvironment (TME) heterogeneously consists of cellular and non-cellular components including the surrounding blood vessels, immune cells, fibroblasts, cancer stem cells and extracellular matrix (ECM).

Figure 2

Tumor-stromal interactions on a chip. (A) 3D Microfluidic model to investigate the carcinoma associated fibroblast promoted tumor spheroid invasion. (i, ii) microfluidic chip design (iii) cell loading step. Salivary gland adenoid cystic carcinoma cell line (ACC-M) were co cultured with carcinoma associated fibroblasts (CAFs). ACC-M invaded CAF-embedded matrix in a spheroid fashion. However, ACC-M cells did not invade the adjacent matrix when co-cultured with the fibroblast cell line (HFL1) [21]; (B) 3D culture of tumor spheroids and fibroblasts in a compartmentalized microfluidic chip. (i, ii) Fluorescence images of HT-29 tumor spheroids and CCD-18Co human normal fibroblast cell line. HT-29 spheroids and CCD-18Co cells proliferated within the space of the corresponding channels over 5 days, during which their growth and interaction were monitored and characterized [26]. Reproduced with permission.

Figure 3

Tumor angiogenesis on a chip. (A) Human glioblastoma multiforme cells, (U87MG) were used to induce angiogenic sprouting. Fluorescence image shows angiogenic sprouts grown for 2 and 4 days under co-culture with U87MG cancer cells and human umbilical vein endothelial cells (HUVEC) (i, ii) [32]. Scale bar: 50 μm; (B) Pre-vascularized tumor (PVT) spheroid model. PVT spheroid model were introduced breast cancer (MCF10A, MDA-MB-231), Lung cancer (A549) and colon cancer (SW620). Representative fluorescence images of PVT spheroid model shows robust angiogenic sprouting. Various PVT spheroid showed different angiogneic sprouting behavior. Intravasation events were only observed for SW620 cancer cells [37]. Scale bar: 100 μm. Reproduced with permission.

Figure 4

Metastasis on a chip. (A) A human 3D vascularized organotypic microfluidic system to study cancer cell extravasation (i) Cancer cell extravasation was monitored in real time within a vascular network (ii) magnified image [43]. Scale bar: 100 μm; (B) Human umbilical vein pericytes were cocultured with human umbilical vein endothelial cells to form pericyte-covered lumens. The extravasation rate from HUVEC-only cultures was significantly higher when compared to HUVEC-pericyte coculture [44]. Scale bar: 20 μm; (C) Design of biomimetic multi-organ chip (i, ii) multi-organ chip included an upstream “lung organ” and three downstream “distant organ” such as bone, brain, liver; (iii, iv) The microfluidic chip was compartmentalized using human epithelial and stromal cells cultured on separated side of a porous membrane in order to mimic (v–vii) physiological respiration in the microfluidic system; which was followed by the introduction of (viii–x) lung fibroblast cells to investigate lung cancer metastasis to distant organ [45]. Reproduced with permission.

Figure 5

Tumor-chemokine interaction on a chip. (A) Chemotaxis in gradients induced cancer cell migration. (i) An in vitro model of tumor-stromal interaction engineerined in a microfluidic chip consisting of porous membrane; (ii) The cellular seeding procedure uses the following color-coded cells: Red (L12) CXCL12 producing cell, Blue(X4) CXCR4 expressing cells; and Green(X7) CXCR7 expressing cells (iii) Time lapse images show progressive migration of X4 cells toward L12 cells [60]. Scale bar: 200 μm; (B) A microfluidic device for study of transendothelial invasion of tumor aggregates by stimulation of chemokine CXCL12. (i, ii) Schematic representation of the device; (iii, iv) Transendothelial invasion of ACC-M aggregates induced by CXCL12. ACC-M aggregates could not transmigrate across HUVEC in the control but ACC-M aggregates transmigrated HUVEC and invaded into ECM when induced by CXCL12 [61]. Reproduced with permission.

Figure 6

Probing the efficacy of drug delivery using TME on a chip. (A) Droplet-based microfluidic system for multicellular tumor spheroid formation and anticancer drug testing. (i) Schematic of the droplet formation and cell culture microfluidic chips. Each chamber contains 14 sieves for alginate droplet trapping; (ii) Breast tumor cells proliferating and forming multicellular spheroids while encapsulated in alginate beads. Tumor cells were perfused with doxorubicin and live/dead assay was assessed [75]. Scale bar: 100 μm; (B) Tumor on a chip provides an optical window into nanoparticle tissue transport. (i) Schematic of the microfluidic device; (ii) MDA-MB-435 breast cancer cell embedded within microfluidic device (iii) Effect of nanoparticle size on tissue accumulation. 40 nm fluorescent PEG-nanoparticles entered the tumor spheroid and accumulated in the interstitial spaces but 110 nm nanoparticles were excluded from the spheroid [93]. Scale bar: 100 μm. Reproduced with permission.