Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 229 of the 231 references it cites were found in a registry or an index; 136 of the 207 sources its claims cite were checked for support.

The Tumor-on-Chip: Recent Advances in the Development of Microfluidic Systems to Recapitulate the Physiology of Solid Tumors

Open access
Trujillo-de Santiago, Grissel, Flores-Garza, Brenda Giselle, Tavares-Negrete, Jorge Alfonso, Lara-Mayorga, Itzel Montserrat, González-Gamboa, Ivonne, Zhang, Yu Shrike, Rojas-Martínez, Augusto, Ortiz-López, Rocío, Álvarez, Mario Moisés · 2019 · Materials

Abstract

The ideal in vitro recreation of the micro-tumor niche-although much needed for a better understanding of cancer etiology and development of better anticancer therapies-is highly challenging. Tumors are complex three-dimensional (3D) tissues that establish a dynamic cross-talk with the surrounding tissues through complex chemical signaling. An extensive body of experimental evidence has established that 3D culture systems more closely recapitulate the architecture and the physiology of human solid tumors when compared with traditional 2D systems. Moreover, conventional 3D culture systems fail to recreate the dynamics of the tumor niche. Tumor-on-chip systems, which are microfluidic devices that aim to recreate relevant features of the tumor physiology, have recently emerged as powerful tools in cancer research. In tumor-on-chip systems, the use of microfluidics adds another dimension of physiological mimicry by allowing a continuous feed of nutrients (and pharmaceutical compounds). Here, we discuss recently published literature related to the culture of solid tumor-like tissues in microfluidic systems (tumor-on-chip devices). Our aim is to provide the readers with an overview of the state of the art on this particular theme and to illustrate the toolbox available today for engineering tumor-like structures (and their environments) in microfluidic devices. The suitability of tumor-on-chip devices is increasing in many areas of cancer research, including the study of the physiology of solid tumors, the screening of novel anticancer pharmaceutical compounds before resourcing to animal models, and the development of personalized treatments. In the years to come, additive manufacturing (3D bioprinting and 3D printing), computational fluid dynamics, and medium- to high-throughput omics will become powerful enablers of a new wave of more sophisticated and effective tumor-on-chip devices.

Overview

CoChat Summary

## TL;DR

This is a tutorial review of tumor-on-chip systems for solid-tumor research, arguing that microfluidic 3D models better recapitulate tumor physiology and dynamics than traditional 2D culture and many animal or static 3D models. It surveys recent device architectures and applications, especially spheroid models, co-cultures with stromal/healthy cells, vascularized chips, drug screening, nanoparticle transport, and precision-medicine uses.

## Key findings

  • The paper states that less than 10% of anticancer drugs entering clinical trials reach the market, motivating more predictive preclinical models.
  • Tumor-on-chip systems are presented as better than 2D culture because they can incorporate 3D architecture, dynamic flow, nutrients, and pharmaceutical compounds, and as more controllable than animal models, which are described as time-, resource-, and expertise-intensive and not fully human-relevant.
  • In a 48-h perfused mesothelioma spheroid chip, spheroid morphology stayed essentially unchanged, but perfused spheroids showed higher chemoresistance to cisplatin than static spheroids; the system used about 5000 cells per assay.
  • Several examples reported quantitative treatment or transport effects, including 100 nm nanoparticles penetrating better than 200 nm particles, DOX at 10 and 100 µg/mL destroying an endothelial barrier in one co-culture model, vessels in breast tumor compartments being 2.62-fold more permeable to nanoparticles than healthy-tissue vessels, and vascularized HCT116 microtumors showing one-order-higher chemoresistance to oxaliplatin than 2D monolayers.

## Methods

  • Design: Narrative/tutorial review organized by increasing model complexity: simple spheroids, spheroids in hydrogels, non-spheroid cancer–normal co-cultures, vascularized systems, and precision-medicine applications.
  • Data/sample: The review focuses on recent literature from 2011 onward on solid-tumor microfluidic systems and cites multiple representative device studies and model systems.
  • Analysis: Comparative synthesis of published examples, emphasizing device architecture, fabrication/materials, flow and pressure control, and reported biological readouts such as drug response, nanoparticle transport, cell invasion, and vascular permeability.

## Limitations

  • The paper states that the complete recapitulation of the true nature and evolution of a malignant tumor in vitro is challenging, and that high-fidelity 3D emulation remains an aspiration.
  • It states that microfluidic device fabrication still requires skill and experience, and that better biomaterials, 3D printing, CFD, and omics will be needed for more advanced systems.
  • Evident from the text: this is a review, not an original experimental study, so its conclusions depend on the quality and scope of the cited literature rather than new primary data.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.3390/ma12182945

Cites

229 of 229 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Figure 1 Recapitulation of the tumor niche using tumor-on-chip systems with various degrees of sophistication. (a) The evolution of our cancer research tools—from 2D to 3D, and beyond—tumor-on-chip systems. Colored circles show the degree of fulfillment of different features. Green: fully fulfilled; Yellow: partially fulfilled; Red: poorly fulfilled; (b) Scheme of the tumor niche. Modified from Han et al., 2016 [19]. (c) Different types of tumor-on-chip systems: from simple spheroids in a microflow to vascularized and perfusable tumor microtissues [20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72].
Figure 1 Recapitulation of the tumor niche using tumor-on-chip systems with various degrees of sophistication. (a) The evolution of our cancer research tools—from 2D to 3D, and beyond—tumor-on-chip systems. Colored circles show the degree of fulfillment of different features. Green: fully fulfilled; Yellow: partially fulfilled; Red: poorly fulfilled; (b) Scheme of the tumor niche. Modified from Han et al., 2016 [19]. (c) Different types of tumor-on-chip systems: from simple spheroids in a microflow to vascularized and perfusable tumor microtissues [20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72].CC BY · PubMed Central
Figure 2 Architecture of tumor spheroids and their fabrication techniques. (a) Schematic representation of the 3D architecture of a tumor spheroid. Modified from Sant and Johnston 2017 [126]; (b) Fabrication of spheroids by the following scaffold-free methods: (i) hanging drop [122,127,128,129], (ii) liquid-overlay method [130,131,132,133,134,135,136], and (iii) force-driven method [129,132,137,138,139,140,141]; (c) Fabrication of spheroids by the following scaffold-based methods: (i) spheroids with a matrix-on top and embedded in a matrix [133,134,135], (ii) spheroids encapsulated within a molded matrix [142,143,144], and (iii) spheroids formed by suspending cancer cells within a liquid matrix and applying microfluidic-based strategies to form spheroids in droplets [119,145,146,147,148].
Figure 2 Architecture of tumor spheroids and their fabrication techniques. (a) Schematic representation of the 3D architecture of a tumor spheroid. Modified from Sant and Johnston 2017 [126]; (b) Fabrication of spheroids by the following scaffold-free methods: (i) hanging drop [122,127,128,129], (ii) liquid-overlay method [130,131,132,133,134,135,136], and (iii) force-driven method [129,132,137,138,139,140,141]; (c) Fabrication of spheroids by the following scaffold-based methods: (i) spheroids with a matrix-on top and embedded in a matrix [133,134,135], (ii) spheroids encapsulated within a molded matrix [142,143,144], and (iii) spheroids formed by suspending cancer cells within a liquid matrix and applying microfluidic-based strategies to form spheroids in droplets [119,145,146,147,148].CC BY · PubMed Central
Figure 3 Simple tumor-on-chip systems: a tumor spheroid continuously perfused within a microfluidic device. (a) Simple serpentine-shaped microfluidic system designed by Ruppen et al. [20], (i) designed to trap cancer cell agglomerates for fabrication of tumor spheroids. A simple hydraulic system enabled the trapping of cell agglomerates (scale bar: 10 μm) (ii) in small cavities. Once the upstream cavities were occupied by spheroids, flow was directed to downstream portions of the serpentine (scale bar: 500 μm) (iii) to favor new trapping events. (iv) Spheroids grew and matured in these cavities and were continuously exposed to anticancer compounds. Taken from reference [20]; (b) Microfluidic device developed by Huang et al. [21]: (i) general architecture, (ii) hydraulic testing, (iii) detail of the spheroid traps, and (iv) actual spheroid trapped in a holder as imaged in bright field. Schematic representation of the effect of (scale bar: 50 μm) (v) protein corona, and (vi) charge, on nanoparticle penetration into tumor spheroids. Taken from reference [21]; (c) Simple pump-independent (i) microfluidic system devised by Patra et al. [22] for continuous formation and perfusion of cancer spheroids (ii) contained within rectangular cavities located at the floor of the microfluidic circuit. (iii) Flow is driven simply by a pressure head induced by a difference in the height of the column of liquid (culture medium) between the inlet and the outlet reservoir. Taken from reference [22]; (d) Tumor-on-chip system developed by Chen et al. [177]: (i) Scheme of the physiological context aimed, (ii) and overall microfluidic architecture. (iii) 3D-render showing cancer cells retained within a horseshoe trap. Red and blue arrows indicate the flow fields within the system. (iv) Scheme of the architecture of the system (side view), (v) detail of the array of traps holding spheroids of consistent size, and (vi) size of spheroids after 14 days of culture. Scale bar: 200 μm. Taken from reference [177]; (e) The microfluidic system developed by Lim and Park [23] enables the exposure of cancer spheroids to anticancer drug gradients. (i) Scheme of the overall architecture of the device, (ii) detail of the cell-culture section, and (iii) close-up showing the geometry of each culture microwell. (iv) Experimental demonstration of the effective generation of gradients using injections of a fluorescent tracer. Taken from reference [23].
Figure 3 Simple tumor-on-chip systems: a tumor spheroid continuously perfused within a microfluidic device. (a) Simple serpentine-shaped microfluidic system designed by Ruppen et al. [20], (i) designed to trap cancer cell agglomerates for fabrication of tumor spheroids. A simple hydraulic system enabled the trapping of cell agglomerates (scale bar: 10 μm) (ii) in small cavities. Once the upstream cavities were occupied by spheroids, flow was directed to downstream portions of the serpentine (scale bar: 500 μm) (iii) to favor new trapping events. (iv) Spheroids grew and matured in these cavities and were continuously exposed to anticancer compounds. Taken from reference [20]; (b) Microfluidic device developed by Huang et al. [21]: (i) general architecture, (ii) hydraulic testing, (iii) detail of the spheroid traps, and (iv) actual spheroid trapped in a holder as imaged in bright field. Schematic representation of the effect of (scale bar: 50 μm) (v) protein corona, and (vi) charge, on nanoparticle penetration into tumor spheroids. Taken from reference [21]; (c) Simple pump-independent (i) microfluidic system devised by Patra et al. [22] for continuous formation and perfusion of cancer spheroids (ii) contained within rectangular cavities located at the floor of the microfluidic circuit. (iii) Flow is driven simply by a pressure head induced by a difference in the height of the column of liquid (culture medium) between the inlet and the outlet reservoir. Taken from reference [22]; (d) Tumor-on-chip system developed by Chen et al. [177]: (i) Scheme of the physiological context aimed, (ii) and overall microfluidic architecture. (iii) 3D-render showing cancer cells retained within a horseshoe trap. Red and blue arrows indicate the flow fields within the system. (iv) Scheme of the architecture of the system (side view), (v) detail of the array of traps holding spheroids of consistent size, and (vi) size of spheroids after 14 days of culture. Scale bar: 200 μm. Taken from reference [177]; (e) The microfluidic system developed by Lim and Park [23] enables the exposure of cancer spheroids to anticancer drug gradients. (i) Scheme of the overall architecture of the device, (ii) detail of the cell-culture section, and (iii) close-up showing the geometry of each culture microwell. (iv) Experimental demonstration of the effective generation of gradients using injections of a fluorescent tracer. Taken from reference [23].CC BY · PubMed Central
Figure 4 Tumor-on-chip in which a spheroid is embedded in a hydrogel. (a) Microfluidic system proposed by Albanese et al. [27]: (i) scheme of the optically accessible TOC, (ii) actual image of the spheroid within the microfluidic chamber (scale bar: 1000 μm). The tumor spheroid is lightly embedded in collagen. Scale bar: 100 μm. (iii) Interstitial volume exchange (residence time within a spheroid) as a function of the inlet flow rate (black squares); the corresponding interstitial flow rate is shown (red diamonds). (iv) Scheme and image of 40 nm PEG nanoparticles penetrating the spheroid and accumulating in the interstitial spaces (scale bar: 100 μm). (v) Larger nanoparticles (d = 110 nm) cannot penetrate the spheroid. Scale bar: 100 µm. (vi) Nanoparticle of different sizes [40 nm (red); 70 nm (blue); 110 (green), and 150 (blue)] exhibit different accumulation rates within spheroids. Nanoparticles were injected into the chamber at 50 mL/h; (b) The tumor on chip designed by Kwak et al. [28] aimed to recapitulate (i) the dynamics induced by the pressure gradient between the capillary and the lymphatic nets. (ii) Scheme (scale bar: 300 μm) and (iii) actual microdevice: cancer spheroids (indicated by arrows) develop within collagen in the central compartment by day 3. These spheroids, like actual tumors, express (iv) E-cadherin and (v) collagen IV. Scale bar: 50 µm. This system enables studies of accumulation of nanoparticles within tumors. (vi) Concentration of 100 nm particles at different distances of the capillary boundary (vii) Accumulation of nanoparticles as a function of matrix stiffness, (viii) interstitial pressure, and (ix) particles size. Taken from reference [28]; (c) TOC devised by Piotrowsky-Daspit [30] to study the effect of intra-tumor interstitial pressure on tumor invasiveness. (i) Schematic representation of the device; (ii) Micrographs showing an E-cadherin-GFP aggregate under a condition where the pressure gradient favors invasiveness (higher pressure at the tumor core). Modified from Piotrowsky-Daspit [30].
Figure 4 Tumor-on-chip in which a spheroid is embedded in a hydrogel. (a) Microfluidic system proposed by Albanese et al. [27]: (i) scheme of the optically accessible TOC, (ii) actual image of the spheroid within the microfluidic chamber (scale bar: 1000 μm). The tumor spheroid is lightly embedded in collagen. Scale bar: 100 μm. (iii) Interstitial volume exchange (residence time within a spheroid) as a function of the inlet flow rate (black squares); the corresponding interstitial flow rate is shown (red diamonds). (iv) Scheme and image of 40 nm PEG nanoparticles penetrating the spheroid and accumulating in the interstitial spaces (scale bar: 100 μm). (v) Larger nanoparticles (d = 110 nm) cannot penetrate the spheroid. Scale bar: 100 µm. (vi) Nanoparticle of different sizes [40 nm (red); 70 nm (blue); 110 (green), and 150 (blue)] exhibit different accumulation rates within spheroids. Nanoparticles were injected into the chamber at 50 mL/h; (b) The tumor on chip designed by Kwak et al. [28] aimed to recapitulate (i) the dynamics induced by the pressure gradient between the capillary and the lymphatic nets. (ii) Scheme (scale bar: 300 μm) and (iii) actual microdevice: cancer spheroids (indicated by arrows) develop within collagen in the central compartment by day 3. These spheroids, like actual tumors, express (iv) E-cadherin and (v) collagen IV. Scale bar: 50 µm. This system enables studies of accumulation of nanoparticles within tumors. (vi) Concentration of 100 nm particles at different distances of the capillary boundary (vii) Accumulation of nanoparticles as a function of matrix stiffness, (viii) interstitial pressure, and (ix) particles size. Taken from reference [28]; (c) TOC devised by Piotrowsky-Daspit [30] to study the effect of intra-tumor interstitial pressure on tumor invasiveness. (i) Schematic representation of the device; (ii) Micrographs showing an E-cadherin-GFP aggregate under a condition where the pressure gradient favors invasiveness (higher pressure at the tumor core). Modified from Piotrowsky-Daspit [30].CC BY · PubMed Central
Figure 5 Tumor-on-chip (TOC) systems in which cancerous and normal microtissues share an interface. (a) Crosstalk between cancer cells and fibroblasts in a Y-shaped TOC: (i) scheme and (ii) actual device. (iii) Human mammary fibroblasts (HMF) and breast cancer (MCF10-DCIS) cells establish a cross-talk at the interface between the cancerous and normal microtissue sections. Scale bar: 30 µm. (iv) Circularity, roundness, and aspect ratio (AR) of MCF10-DCIS cancer cells as a function of their distance from the interface. (v) Image analysis characterization of invasiveness. Taken from reference [33]; (b) On-chip activation of stromal tissue by crosstalk with cancerous tissue [35]: (i) schematic representation of the device and (ii) actual micrographs of normal cells (expressing GFP) and cancer cells (expressing RFP). Scale bar: 100 μm. (iii) Time sequence showing cancer cell migration and invasion. Scale bars: 100 μm; (c) TOC system that recapitulates early stage breast cancer: (i) tumor-niche aimed. (ii) Schematic representation of the TOC system. (iii) Cytotoxicity against cancerous DCIS spheroids and normal epithelial cells at different feeding conditions: continuous administration of Paclitaxel through the lower channel (open triangles), batch administration of paclitaxel on normal epithelial cells (open circles), and negative control (squares). (iv) Change in the projected area of a DCIS spheroid treated with paclitaxel in the TOC (white bars) and untreated (black bars; ** p < 0.05) (v) Micrographs of DCIS spheroids at day 0, after three days of culture, and after three days of paclitaxel treatment. Scale bars: 100 μm; (d) TOC by Ayuso et al. [36] to recapitulate hypoxia and nutrient gradients within cancerous tissue: (i) schematic representation and (ii) actual device. Scale bar: 1 cm. (iii) Evolution of the cancer cell viability profiles, and (iv-v) hypoxia levels within the device. Scale bar: 400 μm Taken from reference [36].
Figure 5 Tumor-on-chip (TOC) systems in which cancerous and normal microtissues share an interface. (a) Crosstalk between cancer cells and fibroblasts in a Y-shaped TOC: (i) scheme and (ii) actual device. (iii) Human mammary fibroblasts (HMF) and breast cancer (MCF10-DCIS) cells establish a cross-talk at the interface between the cancerous and normal microtissue sections. Scale bar: 30 µm. (iv) Circularity, roundness, and aspect ratio (AR) of MCF10-DCIS cancer cells as a function of their distance from the interface. (v) Image analysis characterization of invasiveness. Taken from reference [33]; (b) On-chip activation of stromal tissue by crosstalk with cancerous tissue [35]: (i) schematic representation of the device and (ii) actual micrographs of normal cells (expressing GFP) and cancer cells (expressing RFP). Scale bar: 100 μm. (iii) Time sequence showing cancer cell migration and invasion. Scale bars: 100 μm; (c) TOC system that recapitulates early stage breast cancer: (i) tumor-niche aimed. (ii) Schematic representation of the TOC system. (iii) Cytotoxicity against cancerous DCIS spheroids and normal epithelial cells at different feeding conditions: continuous administration of Paclitaxel through the lower channel (open triangles), batch administration of paclitaxel on normal epithelial cells (open circles), and negative control (squares). (iv) Change in the projected area of a DCIS spheroid treated with paclitaxel in the TOC (white bars) and untreated (black bars; ** p < 0.05) (v) Micrographs of DCIS spheroids at day 0, after three days of culture, and after three days of paclitaxel treatment. Scale bars: 100 μm; (d) TOC by Ayuso et al. [36] to recapitulate hypoxia and nutrient gradients within cancerous tissue: (i) schematic representation and (ii) actual device. Scale bar: 1 cm. (iii) Evolution of the cancer cell viability profiles, and (iv-v) hypoxia levels within the device. Scale bar: 400 μm Taken from reference [36].CC BY · PubMed Central
Figure 6 Tumor-on-chip (TOC) systems that consider spheroids embedded in non-cancerous cell-laden-hydrogels. (a) Chemotaxis-driven assembly of endothelial barrier in a TOC platform [50]. Brightfield images of MCF7 spheroids co-cultured with human umbilical vein endothelial cells (HUVECs) (i) immediately after encapsulation, (ii) after three, and (iii) five days of culture. Scale bars: 200 μm. (iv) Evolution of HUVEC densities within different GelMA hydrogel zones 1 through 5 (zones are indicated on the oval inset). Zones are indicated within the inset. (v) Spheroid size as a function of culture time. Size was quantified as the projected area of the spheroid and normalized to its size at day 0; (*) and (**) indicate statistically significant differences of p < 0.05 and 0.01, respectively in a pairwise t-test. (vi) HUVECs and MCF7 spheroids (as observed in bright field) five (D5) and eight days (D8) after doxorubicin (DOX) treatment. Red arrows indicate the presence (or absence) of endothelial barrier. Scale Bar: 200 μm. (vii) MCF7 spheroid size after treatment with increasing doses of DOX. The spheroid area was normalized with respect to that of untreated spheroids. Significant differences are indicated by (*, **): p < 0.05 and 0.01, respectively (pair wise t-test). Taken from reference [50]; (b) TOC developed by Jeong et al. [51] to study the crosstalk and mutual activation between fibroblasts and cancer cells: (i) schematic of the device, (ii) and (iii) successive close-ups on the channel section, (iv) detail of the cell culture channel, and (iv) comparison of migration distances of fibroblasts in the direction of the cancerous region (blue) and the non-cancerous region (gray). (vi) Micrographs of cancer spheroids treated with different doses of DOX in monoculture and co-culture conditions. Scale bars: 50 μm. (vii) Evaluation of DOX efficacy in co-culture and monoculture conditions in this TOC. Significant differences are indicated by (*): p < 0.05. Taken from reference [51]; (c) TOC developed by Aref et al. [48], to model immune checkpoint blockade. (i) A real tumor is subjected to dissociation (mechanical and enzymatic) to yield dissociated tumor tissue (spheroids, cell agglomerates and single cells); (ii) schematic of Aref’s TOC; (iii) spheroid stained to reveal the presence of calcein AM (green); CD8 T cells (red); tumor cells (EpCAM; purple); and all nucleated cells (Hoechst; blue). Scale bars: 20 μm. (iv) Cell viability, and (v) cytokine expression profile over time in spheroids derived from patients and treated with different immunotherapies: α-PD-1: pembrolizumab, 250 μg mL−1); α -CTLA-4: (ipilimumab, 50 μg mL−1); or a combination. Taken from reference [48]. (d) TOC proposed by Bruce et al. [49] to recreate a bone marrow microenvironment and study acute lymphoblastic leukemia (ALL): (i) schematic representation, (ii) actual image, (iii) and detail of the post array within the device. (iv–vii) Confocal images of co-cultures of SUP-B15 (yellow arrows) and BMSC cells (white arrows) in (iv) 2D static, (v) 3D static, and (scale bars: 500 μm and 100 μm) (vi) 3D dynamic models. Proliferating cell nuclei were stained with Ki67 (green). Actin filaments were stained with phalloidin (red). Cell nuclei were stained with DAPI (blue). Scale bars: 20 μm. (vii) Chemoresistance of tumor cells to Ara-C among tri-culture or monoculture systems, in 2D versus 3D culture, and exposed (or not) to interstitial flow. (+) and (*) indicate significant difference between groups of p < 0.1, and p < 0.05, respectively. Taken from reference [49].
Figure 6 Tumor-on-chip (TOC) systems that consider spheroids embedded in non-cancerous cell-laden-hydrogels. (a) Chemotaxis-driven assembly of endothelial barrier in a TOC platform [50]. Brightfield images of MCF7 spheroids co-cultured with human umbilical vein endothelial cells (HUVECs) (i) immediately after encapsulation, (ii) after three, and (iii) five days of culture. Scale bars: 200 μm. (iv) Evolution of HUVEC densities within different GelMA hydrogel zones 1 through 5 (zones are indicated on the oval inset). Zones are indicated within the inset. (v) Spheroid size as a function of culture time. Size was quantified as the projected area of the spheroid and normalized to its size at day 0; (*) and (**) indicate statistically significant differences of p < 0.05 and 0.01, respectively in a pairwise t-test. (vi) HUVECs and MCF7 spheroids (as observed in bright field) five (D5) and eight days (D8) after doxorubicin (DOX) treatment. Red arrows indicate the presence (or absence) of endothelial barrier. Scale Bar: 200 μm. (vii) MCF7 spheroid size after treatment with increasing doses of DOX. The spheroid area was normalized with respect to that of untreated spheroids. Significant differences are indicated by (*, **): p < 0.05 and 0.01, respectively (pair wise t-test). Taken from reference [50]; (b) TOC developed by Jeong et al. [51] to study the crosstalk and mutual activation between fibroblasts and cancer cells: (i) schematic of the device, (ii) and (iii) successive close-ups on the channel section, (iv) detail of the cell culture channel, and (iv) comparison of migration distances of fibroblasts in the direction of the cancerous region (blue) and the non-cancerous region (gray). (vi) Micrographs of cancer spheroids treated with different doses of DOX in monoculture and co-culture conditions. Scale bars: 50 μm. (vii) Evaluation of DOX efficacy in co-culture and monoculture conditions in this TOC. Significant differences are indicated by (*): p < 0.05. Taken from reference [51]; (c) TOC developed by Aref et al. [48], to model immune checkpoint blockade. (i) A real tumor is subjected to dissociation (mechanical and enzymatic) to yield dissociated tumor tissue (spheroids, cell agglomerates and single cells); (ii) schematic of Aref’s TOC; (iii) spheroid stained to reveal the presence of calcein AM (green); CD8 T cells (red); tumor cells (EpCAM; purple); and all nucleated cells (Hoechst; blue). Scale bars: 20 μm. (iv) Cell viability, and (v) cytokine expression profile over time in spheroids derived from patients and treated with different immunotherapies: α-PD-1: pembrolizumab, 250 μg mL−1); α -CTLA-4: (ipilimumab, 50 μg mL−1); or a combination. Taken from reference [48]. (d) TOC proposed by Bruce et al. [49] to recreate a bone marrow microenvironment and study acute lymphoblastic leukemia (ALL): (i) schematic representation, (ii) actual image, (iii) and detail of the post array within the device. (iv–vii) Confocal images of co-cultures of SUP-B15 (yellow arrows) and BMSC cells (white arrows) in (iv) 2D static, (v) 3D static, and (scale bars: 500 μm and 100 μm) (vi) 3D dynamic models. Proliferating cell nuclei were stained with Ki67 (green). Actin filaments were stained with phalloidin (red). Cell nuclei were stained with DAPI (blue). Scale bars: 20 μm. (vii) Chemoresistance of tumor cells to Ara-C among tri-culture or monoculture systems, in 2D versus 3D culture, and exposed (or not) to interstitial flow. (+) and (*) indicate significant difference between groups of p < 0.1, and p < 0.05, respectively. Taken from reference [49].CC BY · PubMed Central
Figure 7 Vascularized tumor-on-chip (TOC) systems. (a) A gelatin and hyaluronic acid (HA) construct, built in a PDMS device, traversed by an endothelialized microchannel. Taken from reference [54]; (b) TOC devised by Ozkan et al. [55]. (i) Schematic representation of a breast tumor microenvironment connected to a healthy liver niche. (ii) Close-up view of the TOC system. Confocal images show GFP-breast cancer cells (green) and particles circulating inside the vascular channel (red) and FITC-anti-albumin immunostained healthy liver cells (green) and particles circulating inside the vascular channel (red). Scale bar: 500 µm. (iii) CFD simulation showing the cross-sectional shear stress profiles in tumor and healthy vessels. Taken from reference [55]; (c) TOC by Shirure et al. [56]. (i) Schematic representation of the device, and (ii) confocal slice of the microvascular network (green) fabricated within the TOC central chamber and perfused with fluorescently labeled dextran (orange). Scale bar: 50 µm. Taken from reference [56]; (d) Bottom-up engineering of vascularized tumor models proposed in Agarwal et al. [57]. (i) Schematic representation of the strategy. (ii) Comparison of the effect of docetaxel, expressed as IC50, for three different microtumor models: 2D (black), a-vascular 3D (blue) and vascularized 3D (red) models; (*) indicate statistically significant difference of p < 0.05 when 3D vascularized models are compared to 2D culture model, and (#) significant differences of p < 0.05 when 3D vascularized models are compared with 2D and 3D avascular culture models. Taken from reference [57]. (e) Vascularized tumor-on-chip developed by Sobrino et al. [58]. (i) Scheme of the microfluidic device: a PDMS microfluidic chip composed of an array of three tissue chambers, each connected to two straight channels mimicking an arteriole (line of higher pressure) and the venule (line of lower pressure). Channels were also connected to the medium inlets and outlets. A hydrostatic pressure differential of 10 mm H2O was established to enable flow across the microfluidic channel, while a pressure gradient of 5 mm H2O was set across the tissue chambers to favor transport across the hydrogel. (ii) A human-endothelial cell vasculature net was fabricated within diamond-shaped tissue chambers containing EC-like matrix. Remarkably, the development of vasculature in these tissue chambers occurred spontaneously by self-assembly of EC cells initially seeded within the channels and the tissue chamber at day zero. Endothelial cells were continuously transduced with mCherry for easy visualization of the network (iii) Experiments of continuous perfusion of Dextran X show that the vascular network is practically leak-free after several days. (iv) Growth of tumors, fabricated by a spheroid formation method assisted by thrombin, seeded within the vasculature network models after 6, 8, 10, and 12 days of culture at continuous perfusion. (v) Different drugs were tested in this tumor-on-chip system; (vi and vii) the model correctly discriminates between drugs that target the tumor-vasculature formation capabilities and those that target the tumor. Scale bars: 50 µm. Taken from reference [58].
Figure 7 Vascularized tumor-on-chip (TOC) systems. (a) A gelatin and hyaluronic acid (HA) construct, built in a PDMS device, traversed by an endothelialized microchannel. Taken from reference [54]; (b) TOC devised by Ozkan et al. [55]. (i) Schematic representation of a breast tumor microenvironment connected to a healthy liver niche. (ii) Close-up view of the TOC system. Confocal images show GFP-breast cancer cells (green) and particles circulating inside the vascular channel (red) and FITC-anti-albumin immunostained healthy liver cells (green) and particles circulating inside the vascular channel (red). Scale bar: 500 µm. (iii) CFD simulation showing the cross-sectional shear stress profiles in tumor and healthy vessels. Taken from reference [55]; (c) TOC by Shirure et al. [56]. (i) Schematic representation of the device, and (ii) confocal slice of the microvascular network (green) fabricated within the TOC central chamber and perfused with fluorescently labeled dextran (orange). Scale bar: 50 µm. Taken from reference [56]; (d) Bottom-up engineering of vascularized tumor models proposed in Agarwal et al. [57]. (i) Schematic representation of the strategy. (ii) Comparison of the effect of docetaxel, expressed as IC50, for three different microtumor models: 2D (black), a-vascular 3D (blue) and vascularized 3D (red) models; (*) indicate statistically significant difference of p < 0.05 when 3D vascularized models are compared to 2D culture model, and (#) significant differences of p < 0.05 when 3D vascularized models are compared with 2D and 3D avascular culture models. Taken from reference [57]. (e) Vascularized tumor-on-chip developed by Sobrino et al. [58]. (i) Scheme of the microfluidic device: a PDMS microfluidic chip composed of an array of three tissue chambers, each connected to two straight channels mimicking an arteriole (line of higher pressure) and the venule (line of lower pressure). Channels were also connected to the medium inlets and outlets. A hydrostatic pressure differential of 10 mm H2O was established to enable flow across the microfluidic channel, while a pressure gradient of 5 mm H2O was set across the tissue chambers to favor transport across the hydrogel. (ii) A human-endothelial cell vasculature net was fabricated within diamond-shaped tissue chambers containing EC-like matrix. Remarkably, the development of vasculature in these tissue chambers occurred spontaneously by self-assembly of EC cells initially seeded within the channels and the tissue chamber at day zero. Endothelial cells were continuously transduced with mCherry for easy visualization of the network (iii) Experiments of continuous perfusion of Dextran X show that the vascular network is practically leak-free after several days. (iv) Growth of tumors, fabricated by a spheroid formation method assisted by thrombin, seeded within the vasculature network models after 6, 8, 10, and 12 days of culture at continuous perfusion. (v) Different drugs were tested in this tumor-on-chip system; (vi and vii) the model correctly discriminates between drugs that target the tumor-vasculature formation capabilities and those that target the tumor. Scale bars: 50 µm. Taken from reference [58].CC BY · PubMed Central
Figure 8 Application scenarios for tumor-on-chip systems. (a) Fundamental studies [30,31,33,35,36,41,43,44,45,48,50,51,53,54,55,56,60,62,64,65,66,67,68,70,71,72,73,80,85,105,186,187,188,189,190,191,192,193]; (b) Pharmacological screening and testing [21,22,23,27,28,29,32,43,47,50,55,56,57,58,59,63,64,165,177,194]; and (c) personalized/precision medicine strategies [24,37,42,56,69,104,195,196,197,198].
Figure 8 Application scenarios for tumor-on-chip systems. (a) Fundamental studies [30,31,33,35,36,41,43,44,45,48,50,51,53,54,55,56,60,62,64,65,66,67,68,70,71,72,73,80,85,105,186,187,188,189,190,191,192,193]; (b) Pharmacological screening and testing [21,22,23,27,28,29,32,43,47,50,55,56,57,58,59,63,64,165,177,194]; and (c) personalized/precision medicine strategies [24,37,42,56,69,104,195,196,197,198].CC BY · PubMed Central

Verification

Claims the paper makes

103 Valid · 4 Not valid · 30 Not checked · 12 Own results and interpretations

100 more claims · 74 with their cited sources checked

Create a free account to see all 149 claims and to verify the 11 references still open.

See all claims — free account
  • It may state two separate claims as one. 33
  • Probably not a substantial claim. 10
  • The quote only partly states this claim. 2

Create a free account to see these claims.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

1 of 1

Citations

Cites

Show all 231 references

Cited by

See all 184

Metadata

Authors
Trujillo-de Santiago, G., Flores-Garza, B. G., Tavares-Negrete, J. A., Lara-Mayorga, I. M., González-Gamboa, I., Zhang, Y. S., Rojas-Martínez, A., Ortiz-López, R., & Álvarez, M. M.
Affiliations
  • Centro de Biotecnología-FEMSA, Tecnologico de Monterrey, Monterrey, Nuevo León CP 64849, Mexico
  • Tecnológico de Monterrey (MX)
  • Brigham and Women's Hospital (US)
  • Centro de Investigación y Transferencia en Salud, Hospital San José, Tecnologico de Monterrey, Monterrey, Nuevo León CP 64849, Mexico
Corresponding author
Álvarez, Mario Moisés
Venue
Materials
Publisher
MDPI AG
Published
2019-09-11
Volume
12
Issue
18
Pages
2945
ISSN
1996-1944 (ISSN-L)
OpenAlex
W2972592345
Cited by
184
References
231
Citations by year
2026: 12 · 2025: 17 · 2024: 47 · 2023: 21 · 2022: 25 · 2021: 47 · 2020: 13 · 2019: 2
FWCI
1.76
Type
journal article
Language
en
Open access
Locations
  • Materials (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
  • Consejo Nacional de Ciencia y Tecnología (SNI256730, SNI 26048)
  • Tecnologico de Monterrey (002EI CIN01)
Topics
  • Physical Sciences › Engineering › Biomedical Engineering › 3D Printing in Biomedical Research
  • Physical Sciences › Engineering › Biomedical Engineering › Microfluidic and Capillary Electrophoresis Applications
  • Health Sciences › Medicine › Oncology › Cancer Cells and Metastasis
Keywords
tumor-on-a-chip, microfluidics, solid tumors, tumor microenvironment, 3D tumor culture, anticancer drug screening, precision oncology, 3D bioprinting, computational fluid dynamics, high-throughput omics

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: The Tumor-on-Chip: Recent Advances in the Development of Microfluidic Systems to Recapitulate the Physiology of Solid Tumors. https://app.cochat.ai/papers/10.3390/ma12182945
The paper
Trujillo-de Santiago, G., Flores-Garza, B. G., Tavares-Negrete, J. A., Lara-Mayorga, I. M., González-Gamboa, I., Zhang, Y. S., Rojas-Martínez, A., Ortiz-López, R., & Álvarez, M. M. (2019). The Tumor-on-Chip: Recent Advances in the Development of Microfluidic Systems to Recapitulate the Physiology of Solid Tumors. Materials. https://doi.org/10.3390/ma12182945

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.