Trends in

Pharmacological Sciences
Review

Organ-on-a-Chip: A New Paradigm for

Drug Development
Chao Ma,1,2 Yansong Peng,2 Hongtong Li,1 and Weiqiang Chen1,2,3,*

The pharmaceutical industry has been desperately searching for efficient drug Highlights
discovery methods. Organ-on-a-Chip, a cutting-edge technology that can emu- Organ-on-a-Chip is a promising inter-
late the physiological environment and functionality of human organs on a chip disciplinary technique emulating in vivo
physiology and pathology for in vitro
for disease modeling and drug testing, shows great potential for revolutionizing
disease modeling, drug screening, and
the drug development pipeline. However, successful translation of this novel precision medicine.
engineering platform into routine pharmacological and medical scenarios
remains to be realized. In this review, we discuss how the Organ-on-a-Chip tech- The Organ-on-a-Chip technology can
be organically incorporated into the
nology can have critical roles in different preclinical stages of drug development drug development pipeline from early
and highlight the current challenges in translation and commercialization of this drug discovery to preclinical screen-
technology for the pharmacological and medical end-users. Moreover, this re- ing, testing, and translation of new
view sheds light on the future developmental trends and need for a next- drugs, which bridges the gap between
animal studies and clinical trials involving
generation Organ-on-a-Chip platform to bridge the gap between animal studies human subjects.
and clinical trials for the pharmaceutical industry.
The future development of personalized
Organ-on-a-Chip and continuous
Microfluidic Organ-on-a-Chip: A Paradigm Shift in Drug Development integration of novel engineering tools
The pharmaceutical industry has continuously sought a productive and efficient research and (e.g., automation handling, 3D printing.
development (R&D, see Glossary) framework for drug discovery. However, the current in vitro and in situ multisensors) and biological
2D or 3D cell culture and in vivo animal experimentation platforms (Figure 1) remain unsatisfactory concepts (e.g., patient-specific induced
pluripotent stem cells and organoids)
for an efficient and accurate preclinical evaluation of drug efficacy and toxicity before clinical trials into Organ-on-a-Chip platform will un-
can be approved for testing in human subjects [1,2]. To date, animal studies remain the gold stan- precedentedly promote its biomedical
dard for the preclinical validation of drugs in pharmaceutical development; however, the accuracy applications.
and reproducibility of results obtained from animal studies are undermined in humans owing to
species differences between the animal and human systems [3,4]. Given variable responses
and unexpected toxicity in humans, ~40% of newly developed drugs fail clinical trials even after
clearing preclinical evaluation with animal models [5]. Drug development involves the assessment
of the physiological and toxicological effects of numerous compounds and their derivatives
to identify the most effective and safe drug candidates. However, the limitations of low-
throughput in vivo animal studies largely contribute to the prolonged drug development life
cycle and increased development costs. For emergent cases, such as the coronavirus disease
2019 (COVID-19) pandemic, rapid drug-screening platforms are urgently required to accelerate 1
Department of Mechanical and
the development of new therapeutics and vaccines [6]. In addition, although in vitro cell culture in Aerospace Engineering, New York
University, Brooklyn, NY 11201, USA
Petri dishes is a simple and high-throughput method for basic drug screening and testing, 2
Department of Biomedical Engineering,
these cellular models generally lack in vivo tissue microarchitecture and physiological functionality. New York University, Brooklyn, NY
Therefore, alternative tissue models with biomimetic human pathophysiology are urgently required 11201, USA
3
Perlmutter Cancer Center, NYU
to bridge the gap between animal studies and clinical trials involving humans in the drug develop-
Langone Health, New York, NY 10016,
ment pipeline [7]. USA

Recent advances in the microfluidics-based Organ-on-a-Chip technique, also known as

microphysiological systems that mimic the physiology and functionality of human organs on a
chip, have been envisaged to foster a paradigm shift in drug development and personalized *Correspondence:
medicine by replacing animal testing [8]. The origin of the Organ-on-a-Chip can be traced back [email protected] (W. Chen).

Trends in Pharmacological Sciences, February 2021, Vol. 42, No. 2 https://doi.org/10.1016/j.tips.2020.11.009 119
© 2020 Elsevier Ltd. All rights reserved.
 Trends in Pharmacological Sciences

Glossary
3D bioprinting: a biofabrication
strategy that precisely prints bioinks
(e.g., cells, hydrogels, and
biocompatible materials) to reconstruct
the structures and functions of living
systems.
3R principles: replacement, reduction,
and refinement; a guiding principle to
minimize the amount of animal
experimentation worldwide.
Absorption, distribution,
metabolism, and excretion (ADME):
describes the PK of a compound within
an organism.
Blood–brain barrier (BBB): multi-cell
layer structure comprising brain
endothelial cells, astrocytes, and
pericytes; its selective permeability
prevents drugs from nonselectively
entering the brain tissue.
Body-on-a-Chip or Human-on-a-
Chip: type of multi-organ chip that
recapitulates the whole human
physiology within a single platform.
Coronavirus disease 2019
(COVID-19): an infectious disease
caused by severe acute respiratory
Trends in Pharmacological Sciences syndrome coronavirus 2.
Extracellular matrix (ECM): complex
Figure 1. Microfluidic Organ-on-a-Chip platform. Preclinical studies rely on major tools (i.e., 2D or 3D in vitro cell
molecular network of noncellular
cultures, and in vivo animal models) for drug development. 2D in vitro cultures offer a rapid and reproducible way to
components that provides physical
analyze drug responses; however, they lack the 3D physiological tissue environment. Conventional 3D cell cultures in a
support and biochemical/biophysical
hydrogel matrix can provide a 3D culture environment, but still fall short to controllably recapitulate the in vivo physiology
cues for tissue development and
and pathology of the human body. Animal models enable in vivo analysis, yet the species differences between animal and
homeostasis.
human physiological mechanisms and complexity of the in vivo physiology weaken the accuracy and reproducibility of
Induced pluripotent stem cells
experimental results. Microfluidic Organ-on-Chip platforms that enable controllable cell culture within an organotypic
(iPSCs): type of PSC that can be
microarchitectural environment provide a simple, yet more physiologically relevant, platform to controllably and
generated from somatic cells by direct
systematically interrogate human biology. Figure generated using BioRender (BioRender.com).
introduction of four Yamanaka factors
(Myc, Oct3/4, Sox2, and Klf4).
Mechanobiology: an emerging
research field interrogating the
three decades, beginning with the application of microfluidic devices for cell culture and biological
contribution of the hemostasis and
analysis [9–12]. As opposed to tissue engineering, given that it is designed from the viewpoint of dysfunction of mechanical properties of
reductionism, Organ-on-a-Chip does not strive to reproduce the whole tissues or organs at the cells and tissues to maintain cell
original scale for clinical replacement of their human counterparts [12,13]. Instead, this technique function, tissue development, and
pathogenesis.
aims to mimic the key organotypic cellular architecture and functionality, 3D extracellular matrix
Microfluidics: the science and
(ECM), biochemical factors, and biophysical cues at a smaller scale, which serves the purpose technology of manufacturing
for disease modeling and drug screening. As a type of microfluidic device, Organ-on-a-Chip microminiaturized devices for
is fabricated with the silicon-based organic polymer polydimethylsiloxane (PDMS) using manipulating and controlling a small
volume of fluids (typically microliters to
the standard soft lithography technique; as such, the chip has a compact size and
femtoliters) in a network of
microchannels to precisely pattern cells and manipulate various fluidic and chemical parame- microchannels.
ters, such as flow rate, pressure, oxygen, and pH, providing controllable culture conditions Multi-organ chip: type of Organ-on-a-
[9–12] (Box 1). This reflects the in vivo microstructural and functional characteristics of Chip that reproduces organotypic
functions of at least two types of tissues/
human tissues and organs, thus enabling effective and accurate research in medicine, biology, organs to study interorgan reactions.
and pharmacology. Despite the revolution that the Organ-on-a-Chip technique may bring to Organ-on-a-Chip: microfluidic in vitro
the pharmaceutical industry, its overall impact remains to be determined, with challenges culture system upon which single or
existing in the transition from basic research to preclinical integration of this platform into the multiple cell types are controllably
cultured within a 3D extracellular matrix
drug development pipeline.

120 Trends in Pharmacological Sciences, February 2021, Vol. 42, No. 2

Trends in Pharmacological Sciences

Box 1. General Fabrication Processes of Organ-on-a-Chip Devices to recapitulate the physiology and/or
Organ-on-a-Chip is a microfluidic culture system fabricated with PDMS using the soft lithography technique. Since the pathophysiology of in vivo tissues/
PDMS-based replica modeling process (soft lithography) was first developed by the Whitesides group, PDMS polymers organs.
have become one of the main structural materials used in microfluidics due to their high elasticity, gas permeability, optical Organoids-on-a-Chip: conceptual
transparency, and biocompatibility [9–12]. Normally, master molds of the designed structure components are first technology merging organoids with
fabricated following a photolithography-based protocol. Different layers of the microfluidic device are then generated Organ-on-a-Chip to recapitulate the
with PDMS polymers using soft lithography, which replicates the micro-size features from photolithographic molds into complexity of human organs by
PDMS slabs, followed by assembly with glass slides via oxygen plasma-assisted bonding. application of intrinsic tissue
development process and external
PDMS-based microfluidic chips have ease of fabrication, handling, and integration, and enable long-term cell culture, real- engineering method.
time imaging and monitoring of Organ-on-a-Chip cultures [11,12]. The overall steps to manufacture Organ-on-a-Chip Pharmacodynamics (PD):
devices are similar, although different structural designs exist to properly imitate characteristics of different tissues/organs. quantitative relationship between drug
concentrations and biochemical and
physiological responses.
Pharmacokinetics (PK): profiles the
Single- and Multi-Organ-on-a-Chip Systems
dynamic movement of a drug in the
Since the early 2000s, researchers have attempted to apply various microfluidic devices and body over time, such as the kinetics of
lab-on-a-chip systems to enable controllable and organotypic cell culture for in vitro biochemical ADME processes.
and pharmacological analyses [11,12], which incubated the concept of Organ-on-a-Chip system. Photolithography: manufacturing
process that transfers micrometric
In 2010, the Ingber group at the Harvard Medical School reported a Lung-on-a-Chip model built
patterns from a photomask to a light-
on Huh’s early work from the Takayama group [14,15], which attracted attention from both the sensitive chemical photoresist.
biology and engineering communities and was deemed to usher in the blossoming Organ-on-a- Research and development (R&D):
Chip development. By co-culturing lung alveolar and capillary cells on the two sides of a porous process by which the pharmaceutical
industry explores and develops new
membrane in the microfluidic channels of the Lung-on-a-Chip, researchers can study the breathing treatments or drugs.
mechanisms occurring at the alveoli–capillary interface as well as the environmental effects on lung Single-organ chip: type of Organ-on-
cells in vitro, providing a biomimetic model to decipher the pathological mechanisms underlying a-Chip that mainly recapitulates the
various pulmonary or other respiratory diseases, such as COVID-19 [6]. Since then, numerous structure and physiological functions of a
specific tissue or organ.
single-organ chips, such as liver chips [4,16,17], kidney chips [18,19], pancreas chips [20,21], Soft lithography: family of patterning
heart chips [22–26], intestine and gut chips [27–29], blood–brain barrier (BBB) chips techniques to reproduce structures into
[30,31], and bone and bone marrow chips [32–34], have been successfully developed for in- a soft polymer material, mostly PDMS,
from a silicon mold with micro/nanoscale
vestigating disease progression and analyzing adverse drug reactions. These single-organ
features.
chip assays can help identify critical biological mechanisms as well as test drug efficiency
and toxicity in target organs at the preclinical development stage, thus providing a reliable ref-
erence for clinical trials.

While single-organ chips focus on mimicking individual organ functions, multi-organ chips
integrating multiple organ units, such as the gut compartment for drug absorbance, liver
compartment for drug metabolism, and kidney compartment for drug elimination, in a single
chip enable more comprehensive studies [35,36]. For instance, a heart–liver–skin three-
organ system was developed by Pires de Mello et al. [35] to analyze the effects of acute
and chronic drug exposure on both heart and liver functions. In addition, a four-organ chip
integrated with sequentially connected intestine, liver, skin, and kidney compartments, with
stable homeostasis across different organ compartments, was developed for testing the
systemic toxicity of drug candidates [36]. Currently, development of a more advanced
version, termed ‘Body-on-a-Chip’ or ‘Human-on-a-Chip’, is underway to mirror the phys-
iology of the entire human body using a single platform for drug pharmacokinetic (PK) and
pharmacodynamic (PD) analyses [37–39]. For instance, Miller and Shuler [40] developed a
proof-of-concept 13-organ system with various cell lines representing the main parenchymal
organs and physiological barrier tissues of humans, demonstrating a physical framework to
investigate the interorgan commutation in response to drug challenges at the human level.
Undoubtedly, the Human-on-a-Chip platform has the potential to serve as an alternative
model system to replace animal models in drug development, ultimately revolutionizing the
pharmaceutical industry; however, there are numerous technical challenges to be addressed,
given the complexity of the human system.

Trends in Pharmacological Sciences, February 2021, Vol. 42, No. 2 121

 Trends in Pharmacological Sciences

Organ-on-a-Chip Embraces Drug Development: A Perfect Match

The Organ-on-a-Chip technology can be organically incorporated into the drug development
pipeline (Box 2) from early drug discovery to preclinical screening, testing, and translation before
the approval of a drug by the US Food and Drug Administration (FDA)i for use in clinical trials and
finally in the market (Figure 2, Key Figure).

Box 2. Drug Development Pipeline

The entire drug development process ranges from early drug discovery through basic biological research, disease modeling,
and target discovery, preclinical studies with in vitro, ex vivo, and in vivo models, and clinical development with humans
(Phase I, II, and III trials), to finally FDA review (in the USA) and approval, and postmarketing monitoring (Figure I) [64,107].
Preclinical development encompasses activities that link drug discovery in the laboratory to the initiation of clinical trials
involving humans. Typically, preclinical drug development can be divided into three principal stages: (i) early drug discovery
stage: target identification, disease modeling, and drug discovery; (ii) preclinical screening and testing stage: lead optimiza-
tion and PK/PD studies; and (3) preclinical trial and translational stage: validation of drug toxicity and efficacy. Each stage aims
to eliminate drug candidates that are either ineffective or toxic. Specifically, the early drug discovery stage seeks to under-
stand the biological and pathophysiological mechanisms underlying a specific disease and identify druggable targets,
followed by drug modeling and screening to discover candidate compounds. Subsequently, the preclinical screening and
testing stage mainly focuses on the delineation of the PK and PD profiles of the drug candidates using conventional mam-
malian models and aims to determine the initial drug regimens for clinical trials. The preclinical trial and translational stage in-
cludes toxicological and safety studies, since drugs exhibiting no adverse effects in the preclinical stage can lead to hepatic,
cardiac, or neural impairment during clinical trials as well as produce unexpected toxicity induced by drug–drug interactions.
Therefore, the balance between toxicity and efficacy is an important decision-making process during the late preclinical stage
before approving the drug for use in clinical trials involving humans. These clinical data regarding the safety and efficiency of
developed drugs are critical for FDA review and approval; in addition, drugs entering the market will be under surveillance for
any potential adverse drug reactions that were not identified during preclinical and clinical studies.

Trends in Pharmacological Sciences

Figure I. Drug Development Pipeline. The drug development process starts from early drug discovery through basic
biological research, disease modeling, and target discovery to search for potential drugs. These drug candidates identified
during the early stage are then screened and tested with various preclinical studies using in vitro, ex vivo, and in vivo models
before being approved for subsequent human trials, such as Phase I, II, and III trials, during clinical development. Finally, the
developed drug that is determined to be both effective and safe in humans is submitted to the US Food and Drug
Administration (FDA) for regulatory review and commercial approval, after which postmarket measures will be implemented for
monitoring potential adverse drug reactions. Figure generated using BioRender (BioRender.com).

122 Trends in Pharmacological Sciences, February 2021, Vol. 42, No. 2

Trends in Pharmacological Sciences

Key Figure
Organ-on-a-Chip Platforms in Preclinical Drug Development

Trends in Pharmacological Sciences

Figure 2. Preclinical drug development includes three main stages: early drug discovery; preclinical screening and testing; and preclinical trial and translation. (A) A
pancreatic ductal adenocarcinoma (PDAC)-on-a-Chip (left) with a biomimetic vascular network (right) [human umbilical vein endothelial cells (HUVECs) red] and
pancreatic cancer duct (PD7591 cells, green) revealed the Activin-ALK7 pathway as a hypovascularity mechanism for PDAC. (B) A bioengineered glioblastoma brain
tumor model with biomimetic tumor–immune–vascular interactions demonstrated that blockade of immunosuppression contributed by tumor-associated macrophages
improved the response to anti-PD-1 immunotherapy. (C) A nonsmall-cell lung cancer (NSCLC) microenvironment model study found that mechanical forces during lung
breath (vacuum driven in two side channels) may promote dormancy and drug resistance of NSCLC cells. (D) A first-pass multi-organ system was applied to predict
the pharmacokinetic (PK) parameters of nicotine with a linked gut–liver–kidney chip. (E) A liver lobule-on-a-chip comprising a liver cord (green) and a liver sinusoid (red)
was applied to analyze adverse drug reactions induced by unexpected drug–drug interactions. (F) A multi-organ platform integrated with a multiplex biomarker analysis
module was developed to noninvasively monitor liver toxicity, as well as cardiotoxicity mediated by interorgan metabolism. Figure generated using BioRender (BioRender.
com). Adapted, with permission, from [41] (A), [47] (B), [53] (C), [57] (D), [16] (E), and [66] (F).

Early Drug Discovery: Basic Research and Disease Modeling

A central reason for the high failure rate of new drugs in clinical trials is our insufficient understand-
ing of fundamental human pathophysiology and the underlying mechanisms. Organ-on-a-Chip
can better model the human system to pinpoint drug targets in a more controllable and traceable
manner compared with animal models. For instance, the Organ-on-a-Chip platform is a powerful

Trends in Pharmacological Sciences, February 2021, Vol. 42, No. 2 123

 Trends in Pharmacological Sciences

tool for studying the multifaceted processes and mechanisms contributing to cancer progression
and treatment, such as cancer cell migration and invasion, extracellular signaling, biophysical
factors in the tumor microenvironment (TME), and tumor heterogeneity [32,41–45]. In particular,
many 3D cancer-on-a-chip models have been developed to mimic various types of solid and
liquid TMEs involving different stromal components, immune suppressor cells, and chemokines
to elucidate the mechanisms of resistance to chemotherapy and immunotherapy [43,46,47].
An organotypic pancreatic cancer chip model (Figure 2A) was developed to investigate the
interactions of pancreatic ductal adenocarcinoma (PDAC, a major exocrine pancreatic cancer)
with the tumor vascular network, and the activin–ALK7 pathway, identified to be the mechanism
of PDAC hypovascularity, leading to poor drug delivery and chemotherapeutic outcomes of
pancreatic cancer treatment [41]. Using a glioblastoma-on-a-chip brain TME model (Figure 2B),
researchers revealed immunological mechanisms, such as macrophage-mediated angiogenesis
and immunosuppression, underlying the regulation of resistance to both chemotherapy and
immunotherapy [45,47]. In addition to solid tumor models, a 3D B cell acute lymphoblastic leukemia
bone marrow niche model was bioengineered by Ma et al. and Witkowski et al. [46,48] to clarify the
contributions of genetically different leukemia bone marrow niches to chemotherapeutic resistance.
Using this leukemia liquid tumor niche model, the researchers revealed that stromal niche cell-
mediated chemokine CXCL12, vascular cell adhesion protein-1 signal, and leukemia intrinsic
NF-κB signaling, as well as the nonclassical monocytic immune cell subsets, may be key
mediators of, and targets for, regulating the response of leukemia to chemotherapy. More
importantly, multi-organ chip systems linked with the vasculature and circulatory system are
crucial for elucidating local and distant disease development, such as cancer initiation and me-
tastasis [49]. For instance, a four-organ chip that recapitulated lung cancer metastasis to the
brain, bone, and liver revealed tumor-induced tissue damage in the targeted bone and liver
compartments [50].

In addition to revealing the underlying biological signaling and interactions, Organ-on-a-Chip can
be applied to study the contributions of mechanobiological factors (mechanobiology) to disease
progression and therapeutic resistance [32,43,51]. By using the lung chip to construct a nonsmall
cell lung cancer (NSCLC) microenvironment model, Hassell et al. [43] found that mechanical
forces during breathing may promote dormancy and drug resistance of NSCLC cells to tyrosine
kinase inhibitor therapy via the epidermal growth factor receptor and mesenchymal–epithelial
transition protein kinase (Figure 2C). Tumor dormancy and chemoresistance regulated by extra-
cellular physical factors have also been demonstrated in a bone perivascular niche model of me-
tastasized breast cancer, which experienced controllable interstitial flow, oxygen gradient, and
shear stress [32]. Thus, these and other studies have raised confidence in the use of Organ-
on-a-Chip as a replacement of, or an alternative modeling platform to, current animal models to
help uncover critical pathological mechanisms and identify therapeutic biomarkers and targets
for improving disease outcomes.

Preclinical Screening and Testing: Pharmacokinetics and Pharmacodynamics

PK and PD studies follow the identification of drug candidates and targets. Typically, PK studies
profile the drug concentrations at different organ sites during the metabolic processes, termed
the absorption, distribution, metabolism, and elimination (ADME) characterization of a
drug candidate, whereas PD studies examine the effects of the drug (e.g., the relationship
between drug dose and pharmacological or toxicological response) on target organs or tissues.
Multi-organ chip systems comprising major metabolically related organ compartments are partic-
ularly suitable for systematic and in situ PK–PD studies across diverse human organ sites and
under various drug administration conditions [37,52–54]. Of particular interest are the liver and
kidney, which have pivotal roles in drug metabolism, as well as the heart, which is strongly

124 Trends in Pharmacological Sciences, February 2021, Vol. 42, No. 2

Trends in Pharmacological Sciences

affected by drug toxicity. Given the various delivery modes of drugs, such as transdermal delivery,
oral administration, and intravenous injection, the skin, gut, and bone marrow compartments can
also be integrated [37,52]. A PK–PD study using a 3D tumor–liver–bone marrow multi-organ chip
system was first introduced by the Shuler group to assess the toxicity and mechanism of action of
the anticancer drug 5-fluorouracil; the authors found that the liver compartment was more
resistant to the drug than were the bone marrow and tumor compartments [55,56]. Similarly,
gut–liver–kidney and bone marrow–liver–kidney multi-organ systems have been developed by
Herland et al. [57] to predict the PK parameters of nicotine (an orally administered drug for aiding
smoking cessation) and cisplatin (an intravenously injected anticancer drug), such as the maxi-
mum nicotine concentration in the arteriovenous reservoir and the time to reach the maximum
level, which were consistent with clinical data (Figure 2D). The bone marrow–liver–kidney multi-
organ system further confirmed the pharmacological responses to cisplatin: when administered
at a dose of 160 μM for 24 h, cisplatin exhibited no hepatotoxicity in the liver chip, but exhibited
myeloid toxicity and nephrotoxicity in the bone marrow and kidney chips, respectively,
recapturing the in vivo PDs of cisplatin [57]. The consistency of the in vitro quantified and
predicted PK and PD parameters with the corresponding clinical data highlights the high function-
ality of these integrated multi-organ chips, which may help optimize drug regimens for Phase I
clinical trials. Furthermore, a human-on-a-chip platform containing a functional human immune
component (circulating monocytic cells) has been developed to evaluate the tissue-specific immune
responses of the cardiac, skeletal, and hepatic compartments to amiodarone (an anti-arrhythmic
drug), highlighting the potential application of this system to assess tissue-specific responses of a
given drug in the PK–PD profile [58].

To map the processes of drug transport and delivery across different organs on a chip, a fluidic
circulatory system should be established between multiple organ compartments, as well as
modeling the physiological barriers between the vasculature and parenchymal tissues
[49,59,60]. For example, a vascularized multi-organ chip with the intestine, liver, kidney, heart,
lung, skin, BBB, and brain organ compartments fluidically connected via endothelialized vascular
microchannels and periodically perfused with a common blood substitute has been developed
[53]. This platform was successfully applied to capture and quantitatively predict drug distribution
across multi-organ compartments using an inulin tracer. In addition to the vasculature and circu-
latory system, physiological barriers in other tissues or organs, such as BBB [30,31], blood–air
barrier in the lung [14,15], glomerular filtration barrier in the kidney [61], human placental barrier
[62,63] and other tissue–tissue interfaces between the vascular endothelium and the parenchy-
mal cells [64], are also essential for studying drug transport and delivery kinetics. For instance,
the BBB is a highly selective physiological barrier formed by the brain capillary endothelial cells,
astrocytes, and pericytes, which regulates the penetration of biochemical molecules, such as
glucose, from the blood into the brain microenvironment through specific transport proteins
[30,31]. Likewise, either drugs themselves or their metabolites must be assessed during drug
development to ensure that they are either able to penetrate the BBB to treat neural disorders
or be restricted by the BBB to prevent off-target brain damage. A microfluidic BBB model infused
with endothelial cells, neuroblastoma cells, microglia, and astrocytes has been applied to study
organophosphate toxicity to the BBB, such as barrier integrity damage, acetylcholinesterase
inhibition, and cellular viability reduction [31]. Such a BBB platform can be functionally coupled
with chips for other parenchymal organs, such as the jejunum, liver, and kidney, to study the
sequential metabolism of drug across multiple organ compartments and the BBB, highlighting a
systematic platform to validate therapeutics for neural diseases [54]. Overall, it remains technically
challenging to reconstruct and integrate these blood–tissue and tissue–tissue interfaces into multi-
organ chips and, thus, requires a close collaboration among engineers, biologists, pharmacologists,
and computational scientists.

Trends in Pharmacological Sciences, February 2021, Vol. 42, No. 2 125

 Trends in Pharmacological Sciences

Preclinical Trial and Translation: Evaluation of Drug Safety and Efficiency

Many drugs may show no adverse effects on animals during the preclinical stage but unpredict-
ably exhibit hepatic, cardiac, or renal impairments in patients during clinical trials [3,4]. Therefore,
the evaluation of toxicity and efficacy is vital during the late stage of preclinical development and
clinical trials [65]. Thus, human Organ-on-a-Chip can serve as a useful tool for efficient and accu-
rate assessment of drug toxicity before the drug is approved for use in clinical trials. For instance,
a biomimetic human liver chip with lobule-like microarchitectures was used to analyze adverse
reactions induced by drug–drug interactions during liver metabolism, offering a screening
platform for drug toxicity and safety during combinational therapies (Figure 2E) [16]. In addition,
multi-organ chips can be specifically applied to study both on- and off-target effects as well
as interorgan metabolism of drugs [66–68]. For instance, a multi-organ platform can allow the
integration of a multiplex, automated, noninvasive biomarker analysis module to monitor drug tox-
icity in the liver and heart (Figure 2F). The results validated that anticancer drugs, such as
capecitabine, exhibited hepatotoxicity first and as well cardiotoxicity when being metabolized by he-
patocytes into the active form [66]. More recently, McAleer et al. [67] noted variable on- and off-target
effects of anticancer drugs in a multi-organ system, possibly due to liver cell-mediated drug
metabolism. Notably, construction of the multi-organ chips and advanced human-on-a-chip
requires comprehensive and systematic consideration of various biological and technical factors,
such as organ scaling and integration, during conceptualization, but the multidimensionality of
the human body makes this a challenge. Thus, while a complex multi-organ chip can provide a
more physiologically relevant scenario for drug toxicity testing, a balance between feasibility and com-
plexity of the system should be taken into account during development.

Due to genetic and microenvironmental heterogeneities, responses of patients to drugs are

often variable, necessitating the accurate evaluation of therapeutic efficacy and optimization for
individual patients. Organ-on-a-Chip integrates primary cells derived from healthy and patient
donors, demonstrating the feasibility of assessing patient-specific drug responses in an
organotypic human pathophysiological environment [31,69]. For instance, a human small airway
chip infused with cells from patients with chronic obstructive pulmonary disease reliably
reproduced relevant clinical symptoms [69]. Recently, cancer immunotherapies, such as PD-1-
based immune checkpoint blockade and adoptive T cell transfer [i.e., T cell receptor- and chimeric
antigen receptor-engineered T cells (TCR-T and CAR-T)], demonstrated encouraging outcomes in
clinical trials of patients with various cancers [70,71]. However, most patients continued to exhibit
suboptimal responses and even experienced disease relapse, suggesting the need for the stratifi-
cation and selection of patients to achieve effective therapeutic response and disease manage-
ment [72]. Accordingly, 3D glioblastoma chips using different molecular subtypes of patient-
derived tumor cells [47] or organotypic tumor spheroids [73] were engineered to recapture patient
tumor immunity, evaluate patient-specific responsiveness to anti-PD-1 immunotherapy, and
screen for additional therapeutic combinations. Moreover, a micropatterned tumor array was de-
veloped to dynamically monitor CAR-T trafficking and evaluate its killing activity, highlighting signif-
icant differences across CAR-T products of different donors as well as across various CAR-T
constructs, indicating that this chip may serve as a preclinical screening platform for the quality
check of CAR-T products [74]. Taken together, these platforms demonstrate the feasibility of
ex vivo modeling of the tumor immune niche and predict therapeutic efficiency in a given patient,
suggesting the superiority of these models over animal models, which lack human immunity.
Nevertheless, current studies on Organ-on-a-Chips with a human immune component in drug
development remain in their infancy and require significant future research efforts [75–77]. Most
Organ-on-a-Chip systems developed thus far utilize allogeneic cells, and further incorporation of
autologous patient-derived immune and tumor cells is imperative to reliably evaluate and predict
patient outcome during or even before immunotherapy administration in clinical trials.

126 Trends in Pharmacological Sciences, February 2021, Vol. 42, No. 2

Trends in Pharmacological Sciences

Organ-on-a-Chip Marches toward the Market

Major Players, Consumers, and Business Models
Dramatic growth of the market and need for drug development are anticipated once Organ-on-a-
Chip products make their way into the drug discovery segment. Given the emerging Organ-on-a-
Chip market, demands are dynamic and influencing factors are diverse. For example, academic
researchers may require complex microphysiological systems to reveal pathophysiological
mechanisms at the early drug discovery stage; biotech and pharmaceutical companies may
prefer a high-throughput yet low-cost platform for rapid screening of drug candidates, PK–PD
studies, and toxicity and efficiency evaluations at the preclinical drug development stage; and
other end-users, such as clinics, may prefer a standardized system or customized service for
the screening of personalized drugs. Here, we present publicly accessible information on major
players (Table 1) from various sources, including company websites, Crunchbaseii, and other
publications [8,78,79]. At present, most Organ-on-a-Chip start-ups prefer to provide standard
liver and cancer chips because of their high demand in drug PK–PD and toxicity studies at the
preclinical stage. In general, device type and reliability are the contributing factors for market
positioning of Organ-on-a-Chip start-ups [79,80]. Although most current Organ-on-a-Chip
devices share similar fabrication methods, constant upgrading and differentiation of products
while ensuring unique or highly integrated solutions to address the end-users’ needs would
help these start-ups succeed in a competitive market. For instance, Mimetas OrganoPlate®
offers a unique solution with Phaseguides™ for passive liquid handling of cells and gel loading.
This platform is manufactured to support up to 96 tissue models on a single industry-standard
384-well plate to enable compatibility with standard liquid handling and readout equipment,
facilitating high-throughput drug screening. A more integrated solution, the ‘Human Emulation
System,’ comprising organ chips, hardware, and software applications, marketed by Emulate
Inc., offers a highly standardized organ chip platform, which has garnered attention from the
scientific, pharmaceutical, and industrial communities, as well as the venture capital industry.

A clear, business model would help Organ-on-a-Chip start-ups survive in a competitive market.
Currently, three business models exist for Organ-on-a-Chip start-ups: (i) to provide ready-for-cul-
ture microfluidic devices; (ii) to provide fully operational, ready-to-use Organ-on-a-Chip; and (iii) to
offer a holistic full-service solution ranging from the initial design to post-sale training and mainte-
nance [80]. For instance, several versions of HUMIMIC™ chip offered by TissUse allow users to
culture two or more organ compartments of interest to study interorgan communications, and
the ready-for-culture OrganoPlate® from Mimetas has been adapted by several research groups
to establish their own on-chip vascularized barrier tissues, as discussed earlier [30,31]. The gut
model product OrganoReady Caco-2™ from Mimetas is a ready-to-use intestine barrier chip
with biomimetic Caco-2 epithelial tubules. Similar intestine chips are also available from Emulate
Inc., which allow the end-users to perform assays on drug toxicity and transport and study intes-
tinal diseases. In addition to selling devices directly, TissUse offers service contracts to help create
a customized Organ-on-a-Chip platform for drug toxicity evaluation and human disease modeling
through their established proprietary rapid-prototyping procedure. Furthermore, CN-Bio Innova-
tions provides a rapid study service for the on-chip disease modeling of human nonalcoholic
steatohepatitis, validation of disease mechanisms of interest, and therapeutic screening using
their PhysioMimix™ OOC system.

Commercialization of Organ-on-a-Chip
The translation and commercialization of Organ-on-a-Chip emphasize the industrial perspective,
including technology standardization and reliability, ease of operation, cost-effectiveness, and
compliance with government regulations [8,78,79]. Thus, further analytical validations of the
diagnostic and therapeutic effectiveness, reproducibility, and safety of these chips for fulfilling the

Trends in Pharmacological Sciences, February 2021, Vol. 42, No. 2 127

 Trends in Pharmacological Sciences

Table 1. Examples of Organ-on-a-Chip Start-ups Worldwide

Company Found University spin-off Scientific founders Major products Region
Year
HμREL 2006 Cornell University Greg Baxter, Robert Freedman Liver chip USA
Kirkstall 2006 University of Pisa John Wilkinson Quasi Vivo® system UK
Hepregen, 2007 MIT Sangeeta Bhatia Liver, islet, cancer model, accessory USA
(merged with BioIVT) devices
CN-Bio Innovations 2009 MIT Linda G. Griffith Liver, gut, skin, heart, lung, kidney, UK
brain chips
InSphero 2009 N/A Jan Lichtenberg, Jens M. Kelm, Liver, islet, tumor cell culture Switzerland
Wolfgang Moritz
TissUse 2010 Berlin Institute of Uwe Marx Multi-Organ-on-a-Chip, accessory Germany
Technology devices
Nortis 2012 Washington University Thomas Neumann Kidney, Liver, USA
Multi-Organ-on-a-Chip, accessory
devices
Emulate 2013 Harvard University Donald Ingber Liver, kidney, lung, intestine chips, USA
accessory devices
Mimetas 2013 Leiden University Jos Joore, Paul Vulto, Thomas Kidney, gut, tumors, liver, lung, The Netherlands
Hankermeier intestine, blood vessel, neuronal
models, accessory devices
AxoSim 2014 Tulane University Michael Moore Nerve-on-chip USA
SynVIVO 2014 N/A Kapil Pant, B. Prabhakar SynTumor, SynBBB, SynRAM, USA
Pandian SynTox
Tara Biosystems 2014 Toronto University Milica Radisic, Gordana Biowire™ II platform USA
Vunjak-Novakovic, Robert
Langer, John M. Baldoni
Alveolix 2015 University of Bern Olivier Guenat Lung-on-chip Switzerland
ANANDA Devices 2015 N/A Margaret Magdesian Neuro device Canada
Hesperos 2015 Cornell University, Michael Shuler, James Hickman Heart, liver, lung, brain, skin, kidney USA
Central Florida University chips, Multi-Organ-on-a-Chip
Altis BioSystems 2016 University of North Nancy Allbritton RepliGut Kits USA
Carolina, Chapel Hill
MesoBioTech 2016 N/A Yong Chen Microfluidics, lung chip France
BiomimX 2017 Polytechnic University Alberto Redaelli Heart-on-a-chip Italy
of Milan
BI/OND 2017 Delft University of Cinzia Silvestri Organoids cultivation, tissue–tissue The Netherlands
Technology interface
Jiksak 2017 N/A Jiro Kawada, Keita Shibuya, Nerve organoids Japan
Bioengineering Norihiro Yumoto, Shinji Tokunaga
DAXIANG 2018 N/A Yu Zhou Liver chip, cancer chip China
Aracari Bio 2019 University of California, G. Wesley Hatfield, Christopher Vascularized micro-organ chip USA
Irvine C.W. Hughes, Steven C.
George, Abraham P. Lee
REVIVO Biosystems 2019 Agency for Science, Massimo Alberti Microfluidic Skin-on-a-Chip Singapore
Technology and Research

practical needs of pharmacological and medical application, as well as the FDA regulations, are
warranted. Early and close collaboration between academic institutes, industrial R&D departments,
and healthcare agencies during each stage of Organ-on-a-Chip development will fulfill different
interests and needs and generate a positive feedback loop to corroborate the effectiveness of
Organ-on-a-Chip platforms and maximize their utility in the actual healthcare industry. For instance,

128 Trends in Pharmacological Sciences, February 2021, Vol. 42, No. 2

Trends in Pharmacological Sciences

Emulate Inc. has partnered with AstraZeneca, Johnson & Johnson, Merck, Takeda, and FDA for
the effectiveness validation of their various products by assessing the safety and efficacy of the
drug candidates for human use in an industrial environment.

Another commercialization challenge results from the limited venture capital investment in the
organ chip field despite the huge potential of this technology, with only a few Organ-on-a-Chip
start-ups receiving substantial investments, such as Emulate Inc. (US$142.3 million), InSphero
(US$35.2 million), and Mimetas (US$34.2 million), according to public information on Crunchbase.
National healthcare and research systems can have a vital role in creating partnerships with
academic institutes and companies, supporting proprietary protection and providing funding
opportunities. For instance, the US National Center for Advancing Translational Science has
cooperated with other agencies to launch a series of Organ-on-a-Chip programs to foster the
use of this technology for practical drug development. US federal agencies, such as the National
Institute of Health, National Science Foundation, and Department of Defense, provide seed
funds through the Small Business Innovation Research and Small Business Technology Transfer
programs to foster the development, standardization, and commercialization of the Organ-on-a-
Chip technology for use in the drug development pipeline. Outside the USA, several universities
and research centers across Europe initiated an open project (Organ-on-Chip in Development,
ORCHIDiii) in 2017 and subsequently established the European Organ-on-Chip Society (EUROoCSiv)
to facilitate a collaborative network across academic, research, industrial, and regulatory institutions
to advance the Organ-on-a-Chip technology and its general application. Notably, the Asia-Pacific
region (primarily China, Singapore, South Korea, and Japan) is deemed to be the emerging market
owing to government support for healthcare technologies. For instance, the Chinese Academy of
Science introduced a 5-year initiative of ‘Organ Reconstruction and Manufacturing’v in 2018. With
consistently increasing investment, the technologies are expected to continue to improve while the
cost continues to decrease. Market needs for Organ-on-a-Chip products are anticipated to grow
rapidly and to be well accepted by the pharmaceutical industry.

Future Advances in Organ-on-a-Chip: Challenges and Opportunities

The development of Organ-on-a-Chip platforms over the past two decades has demonstrated
their potential as new tools for drug discovery and development. New Organ-on-a-Chip platforms
with significant improvements in functionality, integration, automation, manufacture, and person-
alized precision medicine have begun to emerge to meet the growing need for better preclinical
models for drug development.

First, future Organ-on-a-Chip platforms will demonstrate a physiologically relevant and spatio-
temporally responsive microenvironment for solving biological and pharmaceutical problems of
interest. The technological functions of future Organ-on-a-Chip platforms will allow for real-
time, in situ, and dynamic maintenance and monitoring of a large array of biological parameters,
such as shear stress, pH, oxygen, cytokines, and chemokines, as well as downstream and off-
chip analyses of molecular signature, cellular physiology, and tissue pathology, using traditional
analytical tools, such as ELISA, PCR, and single-cell mRNA sequencing (scRNA-seq). Such a
series of functionalities can be achieved with more advanced incorporation of in situ optical,
electrical, chemical, and biological biosensors as integrative microfluidic components to detect
key signals in a spatiotemporal manner, which is challenging in animal models [25,66,81,82].
For instance, the integration of label-free and multiplexed nanoplasmonic sensors for in situ
analysis of cytokine secretion during drug and proinflammatory stimulation in a biomimetic
microfluidic adipose-tissue-on-a-chip platform highlighted its potential as a high-throughput
and an integrated preclinical readout system for drug testing [82]. Development and future
integration of new multi-omics-detecting approaches, such as deterministic barcoding in tissue

Trends in Pharmacological Sciences, February 2021, Vol. 42, No. 2 129

 Trends in Pharmacological Sciences

for spatial omics sequencing (DBiT-seq), in microfluidic Organ-on-a-Chip platforms will allow for
spatial barcoding and sequencing of large amounts of molecular information from tissues at a
genome-scale resolution [83]. We believe that integration with novel concepts and technologies
will continuously improve the performance of Organ-on-a-Chip, endorsing its wider application
in drug development.

Second, future Organ-on-a-Chip platforms require improvement and standardization of the

product manufacturing process as well as categorizing of the system designs, configurable
modules, and interfaces. Currently, most Organ-on-a-Chip devices are manually fabricated
with PDMS in research laboratories using soft lithography. The throughput and reproducibility
of fabrication are questionable for the large-scale production of devices for the market, necessi-
tating a standardized and high-throughput yet low-cost manufacturing process. Implementation
of advanced additive manufacturing methods (e.g., 3D bioprinting) or current standard
manufacturing materials and methods (e.g., injection modeling and laser cutting) as well as the
use of a more standardized, modular format with biologically inert materials (e.g., plastics to
replace PDMS) should be considered [39]. For example, 3D bioprinting is a promising method
for fabricating Organ-on-a-Chip devices, for which sophisticated tissue architectures, complex
scaffolds, or templates of a device can be programmed in advance and automatically printed
with high fidelity and controllability [81,84–86]. By offering a one-step approach of tissue recon-
struction and culture platform engineering, 3D bioprinting and other additive manufacturing
methods are expected to extensively transform the Organ-on-a-Chip fabrication protocol in the
near future. Moreover, the operation of Organ-on-a-Chip platforms should occur in a more auto-
mated, high-throughput and parallelized manner through a standardized user-friendly interface to
enable compatibility with routine biological laboratory experiments and work mode of the phar-
maceutical industry [53,87,88]. Early attempts to use robotic interrogators have shown some
success, such as automated cell culture, intercompartmental fluidic coupling, repeated sample
collection, and in situ microscopic imaging for weeks, in an integrated eight-organ chip [53].
Most recently, a microfluidic platform, IFlowPlate, built on a 384-well plate, enabled in vitro
perfusable culture and vascularization of patient-derived colon organoids, providing a higher
throughput capacity compared with current organ chips [88]. Such upgrades of the Organ-on-
a-Chip platforms will be critical for their commercialization and promotion for acceptance by
the end-users.

Last but not the least, future Organ-on-a-Chip platforms will be developed based on patient-
derived materials, such as patient tissue, decellularized ECM, and other biological materials, for
personalized precision medicine, in which patient selection and stratification biomarkers will be
critical factors leading to successful drug development [72]. In many cases, patient tissue cells
are either limited in number and have low proliferative potential or invasive sample collection is
required; such unavailability and unreliability of patient cell sources present a significant barrier.
For instance, lack of functional human podocytes can hinder the on-chip structural formation of
glomerulus with selective filtration characteristics [18,19], as also evidenced for human neural
and cardiac chips [89–92]. Recent studies have shown that patient induced pluripotent stem
cells (iPSCs), such as those generated from skin-derived fibroblasts, are an alternative and unlim-
ited cell source to produce autologous target organs or tissues, thus enabling the construction of
patient-specific organ chips for personalized disease modeling and drug screening [93]. For
instance, a human BBB chip constructed with patient iPSC-derived neurons, astrocytes, and
brain microvascular endothelial-like cells demonstrated patient-specific disruption of barrier integrity
and blood-to-brain permeability of pharmacologics [94]. Additionally, using healthy or patient
iPSC-derived cardiomyocytes, researchers achieved in vitro modeling of cardiovascular
diseases, screening of potential drugs, and evaluation of cardiac toxicity resulting from drug

130 Trends in Pharmacological Sciences, February 2021, Vol. 42, No. 2

Trends in Pharmacological Sciences

interactions and nanomaterials, all of which could not previously be achieved because of the Outstanding Questions
lack of human cardiac cells [95–100]. Furthermore, the integration of iPSC-based organoids, What are the existing challenges
another concept in which an ex vivo organotypic microtissue is developed via self- preventing the integration of Organ-
on-a-Chip systems into the drug
organization and differentiation of stem cells in a 3D matrix, into the Organ-on-a-Chip platform
development pipeline?
resulted in the construction of a powerful hybrid tool, ‘Organoids-on-a-Chip’ [101–104]. For
instance, microfluidic kidney and retinal organoid chips have been demonstrated to be more How can Organ-on-a-Chip systems
physiologically relevant in terms of tissue maturity and functionality [103,104]. Although it accurately reproduce human immunol-
ogy and discern ‘self’ from ‘non-self’ to
was not possible to discuss all the relevant studies here because of space constraints, attempts screen and assess novel immunother-
to develop patient iPSC-derived organ chips could overcome the inadequacy of traditional ‘one- apeutics, such as immune checkpoint
size-fits-all’ therapeutics, providing an ideal treatment for individual patients across large popula- blockade and CAR-T immunotherapy?
tions for the same disorder. Furthermore, several future advantages are envisioned, particularly in
How can Organ-on-a-Chip systems
the modeling and analysis of rare human diseases, which are restricted by available biological stud- become preclinical and/or clinical tools
ies and subsequent high R&D costs [105,106]. for personalized precision medicine
for patient selection and stratification,
as well as screening of customized
Concluding Remarks therapeutics?
Although there is a need for a paradigm shift in drug development to improve the overall pass rate
of newly developed drugs, the industrial drug development process is relatively standardized [107]. How can Organ-on-a-Chip platforms
outcompete conventional animal ex-
Since the entire drug development process may involve going back and forth from disease
perimentations for preclinical drug
modeling to drug testing, the design and incorporation of the Organ-on-a-Chip platforms into the development and thereby fulfill the
drug development pipeline could be beneficial at all preclinical stages and even realize trial-on- 3R principles?
chips for clinical validation [108]. Given the increasing concerns over animal welfare and rights in
biological experiments, Organ-on-a-Chip platforms can be a promising alternative to avoid ethical
issues related to animal use, and live up to the guiding 3R principles (i.e., replacement, reduction,
and refinement) [109]. However, Organ-on-a-Chip systems are still marginalized in the pharmaceu-
tical industry owing to current challenges to meeting the practical needs of rapid drug discovery and
accurate preclinical evaluation (see Outstanding Questions). From a long-term viewpoint, continued
integration of novel concepts and techniques into the Organ-on-a-Chip platform is expected to
bridge biological and technical gaps between translational, preclinical, and clinical studies. Thus,
in summary, we are enthusiastic in terms of the potential of Organ-on-a-Chip use in the pharma-
ceutical industry and its increasingly promising future in personalized precision medicine.

Acknowledgment
This work was supported by the US National Institutes of Health (R21EB025406 and R35GM133646).

Resources
i
www.fda.gov/home
ii
www.crunchbase.com
iii
https://h2020-orchid.eu/
iv
www.euroocs.eu
v
www.cas.cn

References
1. Cook, D. et al. (2014) Lessons learned from the fate of 6. Si, L. et al. (2020) Human organs-on-chips as tools for
AstraZeneca's drug pipeline: a five-dimensional framework. repurposing approved drugs as potential influenza and
Nat. Rev. Drug Discov. 13, 419–431 COVID19 therapeutics in viral pandemics. bioRxiv Published
2. Dugger, S.A. et al. (2018) Drug development in the era of online August 19, 2020. https://doi.org/10.1101/2020.04.
precision medicine. Nat. Rev. Drug Discov. 17, 183–196 13.039917
3. Day, C.P. et al. (2015) Preclinical mouse cancer models: a 7. Franzen, N. et al. (2019) Impact of organ-on-a-chip technology
maze of opportunities and challenges. Cell 163, 39–53 on pharmaceutical R&D costs. Drug Discov. Today 24,
4. Jang, K.J. et al. (2019) Reproducing human and cross-species 1720–1724
drug toxicities using a Liver-Chip. Sci. Transl. Med. 11, eaax5516 8. Zhang, B. et al. (2018) Advances in organ-on-a-chip engineering.
5. Van Norman, G.A. (2019) Limitations of animal studies for Nat. Rev. Mater. 3, 257–278
predicting toxicity in clinical trials: is it time to rethink our 9. Xia, Y. and Whitesides, G.M. (1998) Soft lithography. Angew.
current approach? JACC Basic Transl. Sci. 4, 845–854 Chem. Int. Ed. Eng. 37, 550–575

Trends in Pharmacological Sciences, February 2021, Vol. 42, No. 2 131

 Trends in Pharmacological Sciences

10. Duffy, D.C. et al. (1998) Rapid prototyping of microfluidic 37. Abaci, H.E. and Shuler, M.L. (2015) Human-on-a-chip
systems in poly(dimethylsiloxane). Anal. Chem. 70, 4974–4984 design strategies and principles for physiologically based
11. Whitesides, G.M. (2006) The origins and the future of pharmacokinetics/pharmacodynamics modeling. Integr.
microfluidics. Nature 442, 368–373 Biol. (Camb) 7, 383–391
12. Bhatia, S.N. and Ingber, D.E. (2014) Microfluidic organs-on- 38. Zhang, C. et al. (2009) Towards a human-on-chip: culturing
chips. Nat. Biotechnol. 32, 760–772 multiple cell types on a chip with compartmentalized
13. Langer, R. and Vacanti, J. (1993) Tissue engineering. Science microenvironments. Lab Chip 9, 3185–3192
260, 920–926 39. Sung, J.H. et al. (2019) Recent advances in body-on-a-chip
14. Huh, D. et al. (2007) Acoustically detectable cellular-level lung injury systems. Anal. Chem. 91, 330–351
induced by fluid mechanical stresses in microfluidic airway 40. Miller, P.G. and Shuler, M.L. (2016) Design and demonstration
systems. Proc. Natl. Acad. Sci. U. S. A. 104, 18886–18891 of a pumpless 14 compartment microphysiological system.
15. Huh, D. et al. (2010) Reconstituting organ-level lung functions Biotechnol. Bioeng. 113, 2213–2227
on a chip. Science 328, 1662–1668 41. Nguyen, D.T. et al. (2019) A biomimetic pancreatic cancer on-
16. Ma, C. et al. (2016) On-chip construction of liver lobule-like chip reveals endothelial ablation via ALK7 signaling. Sci. Adv. 5,
microtissue and its application for adverse drug reaction eaav6789
assay. Anal. Chem. 88, 1719–1727 42. Xiao, Y. et al. (2019) Ex vivo dynamics of human glioblastoma
17. Ma, C. et al. (2016) Pneumatic-aided micro-molding for flexible cells in a microvasculature-on-a-chip system correlates with
fabrication of homogeneous and heterogeneous cell-laden tumor heterogeneity and subtypes. Adv. Sci. (Weinh) 6,
microgels. Lab Chip 16, 2609–2617 1801531
18. Mu, X. et al. (2013) Engineering a 3D vascular network in 43. Hassell, B.A. et al. (2017) Human organ chip models recapitu-
hydrogel for mimicking a nephron. Lab Chip 13, 1612–1618 late orthotopic lung cancer growth, therapeutic responses, and
19. Musah, S. et al. (2017) Mature induced-pluripotent-stem-cell- tumor dormancy in vitro. Cell Rep. 21, 508–516
derived human podocytes reconstitute kidney glomerular- 44. Toh, Y.C. et al. (2018) A 3D microfluidic model to recapitu-
capillary-wall function on a chip. Nat. Biomed. Eng. 1, 0069 late cancer cell migration and invasion. Bioengineering (Basel)
20. Shik Mun, K. et al. (2019) Patient-derived pancreas-on-a-chip 5, 29
to model cystic fibrosis-related disorders. Nat. Commun. 10, 45. Cui, X. et al. (2018) Hacking macrophage-associated immu-
3124 nosuppression for regulating glioblastoma angiogenesis.
21. Glieberman, A.L. et al. (2019) Synchronized stimulation and Biomaterials 161, 164–178
continuous insulin sensing in a microfluidic human Islet on a 46. Ma, C. et al. (2020) Leukemia-on-a-chip: dissecting the
Chip designed for scalable manufacturing. Lab Chip 19, chemoresistance mechanisms in B cell acute lymphoblastic
2993–3010 leukemia bone marrow niche. Sci. Adv. 6, eaba5536
22. Ahn, S. et al. (2018) Mussel-inspired 3D fiber scaffolds for 47. Cui, X. et al. (2020) Dissecting the immunosuppressive tumor
heart-on-a-chip toxicity studies of engineered nanomaterials. microenvironments in Glioblastoma-on-a-Chip for optimized
Anal. Bioanal. Chem. 410, 6141–6154 PD-1 immunotherapy. eLife 9, e52253
23. Marsano, A. et al. (2016) Beating heart on a chip: a novel 48. Witkowski, M.T. et al. (2020) Extensive remodeling of the
microfluidic platform to generate functional 3D cardiac immune microenvironment in B cell acute lymphoblastic
microtissues. Lab Chip 16, 599–610 leukemia. Cancer Cell 37, 867–882
24. Ugolini, G.S. et al. (2018) Generation of functional cardiac 49. Sontheimer-Phelps, A. et al. (2019) Modelling cancer in
microtissues in a beating heart-on-a-chip. Methods Cell Biol. microfluidic human organs-on-chips. Nat. Rev. Cancer 19,
146, 69–84 65–81
25. Zhang, X. et al. (2016) High-throughput assessment of drug 50. Xu, Z. et al. (2016) Design and construction of a multi-organ
cardiac safety using a high-speed impedance detection microfluidic chip mimicking the in vivo microenvironment of
technology-based heart-on-a-chip. Micromachines (Basel) 7, lung cancer metastasis. ACS Appl. Mater. Interfaces 8,
122 25840–25847
26. Sheehy, S.P. et al. (2017) Toward improved myocardial maturity 51. Asmani, M. et al. (2018) Fibrotic microtissue array to predict
in an organ-on-chip platform with immature cardiac myocytes. anti-fibrosis drug efficacy. Nat. Commun. 9, 2066
Exp. Biol. Med. (Maywood) 242, 1643–1656 52. Prantil-Baun, R. et al. (2018) Physiologically based pharmacokinetic
27. Shim, K.Y. et al. (2017) Microfluidic gut-on-a-chip with three- and pharmacodynamic analysis enabled by microfluidically linked
dimensional villi structure. Biomed. Microdevices 19, 37 organs-on-chips. Annu. Rev. Pharmacol. Toxicol. 58, 37–64
28. Poceviciute, R. and Ismagilov, R.F. (2019) Human-gut- 53. Novak, R. et al. (2020) Robotic fluidic coupling and interroga-
microbiome on a chip. Nat. Biomed. Eng. 3, 500–501 tion of multiple vascularized organ chips. Nat. Biomed. Eng.
29. Guo, Y. et al. (2018) A biomimetic human gut-on-a-chip for 4, 407–420
modeling drug metabolism in intestine. Artif. Organs 42, 54. Vernetti, L. et al. (2017) Functional coupling of human
1196–1205 microphysiology systems: intestine, liver, kidney proximal
30. Wevers, N.R. et al. (2018) A perfused human blood-brain tubule, blood-brain barrier and skeletal muscle. Sci. Rep. 7,
barrier on-a-chip for high-throughput assessment of barrier 42296
function and antibody transport. Fluids Barriers CNS 15, 23 55. Sung, J.H. and Shuler, M.L. (2009) A micro cell culture analog
31. Koo, Y. et al. (2018) Three-dimensional (3D) tetra-culture brain (microCCA) with 3-D hydrogel culture of multiple cell lines to
on chip platform for organophosphate toxicity screening. Sci. assess metabolism-dependent cytotoxicity of anti-cancer
Rep. 8, 2841 drugs. Lab Chip 9, 1385–1394
32. Marturano-Kruik, A. et al. (2018) Human bone perivascular 56. Sung, J.H. et al. (2010) A microfluidic device for a
niche-on-a-chip for studying metastatic colonization. Proc. pharmacokinetic-pharmacodynamic (PK-PD) model on a
Natl. Acad. Sci. U. S. A. 115, 1256–1261 chip. Lab Chip 10, 446–455
33. Hao, S. et al. (2018) A spontaneous 3D bone-on-a-chip for 57. Herland, A. et al. (2020) Quantitative prediction of human
bone metastasis study of breast cancer cells. Small 14, pharmacokinetic responses to drugs via fluidically coupled
e1702787 vascularized organ chips. Nat. Biomed. Eng. 4, 421–436
34. Torisawa, Y.S. et al. (2016) Modeling hematopoiesis and 58. Sasserath, T. et al. (2020) Differential monocyte actuation in a
responses to radiation countermeasures in a bone marrow- three-organ functional innate immune system-on-a-chip. Adv.
on-a-chip. Tissue Eng. Part C Methods 22, 509–515 Sci. (Weinh) 7, 2000323
35. Pires de Mello, C.P. et al. (2020) Microphysiological heart-liver 59. Park, D. et al. (2020) Integrating organs-on-chips: multiplexing,
body-on-a-chip system with a skin mimic for evaluating topical scaling, vascularization, and innervation. Trends Biotechnol.
drug delivery. Lab Chip 20, 749–759 38, 99–112
36. Maschmeyer, I. et al. (2015) A four-organ-chip for intercon- 60. van der Helm, M.W. et al. (2016) Microfluidic organ-on-chip
nected long-term co-culture of human intestine, liver, skin technology for blood-brain barrier research. Tissue Barriers 4,
and kidney equivalents. Lab Chip 15, 2688–2699 e1142493

132 Trends in Pharmacological Sciences, February 2021, Vol. 42, No. 2

Trends in Pharmacological Sciences

61. Qu, Y.Y. et al. (2018) A nephron model for study of drug- 86. Yi, H.G. et al. (2019) A bioprinted human-glioblastoma-on-a-
induced acute kidney injury and assessment of drug-induced chip for the identification of patient-specific responses to
nephrotoxicity. Biomaterials 155, 41–53 chemoradiotherapy. Nat. Biomed. Eng. 3, 509–519
62. Blundell, C. et al. (2016) A microphysiological model of the 87. Benam, K.H. et al. (2016) Matched-comparative modeling of
human placental barrier. Lab Chip 16, 3065–3073 normal and diseased human airway responses using a
63. Yin, F.C. et al. (2019) A 3D human placenta-on-a-chip model to microengineered breathing lung chip. Cell Syst. 3, 456–466
probe nanoparticle exposure at the placental barrier. Toxicol. in 88. Rajasekar, S. et al. (2020) IFlowPlate-A customized 384-well
Vitro 54, 105–113 plate for the culture of perfusable vascularized colon
64. Sakolish, C.M. et al. (2016) Modeling barrier tissues in vitro: organoids. Adv. Mater. Published online October 1, 2020.
methods, achievements, and challenges. EBioMedicine 5, https://doi.org/10.1002/adma.202002974
30–39 89. Yi, Y. et al. (2015) Central nervous system and its disease
65. Cirit, M. and Stokes, C.L. (2018) Maximizing the impact of models on a chip. Trends Biotechnol. 33, 762–776
microphysiological systems with in vitro-in vivo translation. 90. Mathur, A. et al. (2015) Human iPSC-based cardiac
Lab Chip 18, 1831–1837 microphysiological system for drug screening applications.
66. Zhang, Y.S. et al. (2017) Multisensor-integrated organs-on- Sci. Rep. 5, 8883
chips platform for automated and continual in situ monitoring 91. Zhao, Y. et al. (2019) A platform for generation of chamber-
of organoid behaviors. Proc. Natl. Acad. Sci. U. S. A. 114, specific cardiac tissues and disease modeling. Cell 176,
E2293–E2302 913–927
67. McAleer, C.W. et al. (2019) Multi-organ system for the evaluation 92. Motallebnejad, P. et al. (2019) An isogenic hiPSC-derived
of efficacy and off-target toxicity of anticancer therapeutics. Sci. BBB-on-a-chip. Biomicrofluidics 13, 064119
Transl. Med. 11, eaav1386 93. Liu, L.P. et al. (2019) Therapeutic potential of patient iPSC-
68. Esch, E.W. et al. (2015) Organs-on-chips at the frontiers of derived iMelanocytes in autologous transplantation. Cell Rep.
drug discovery. Nat. Rev. Drug Discov. 14, 248–260 27, 455–466
69. Benam, K.H. et al. (2016) Small airway-on-a-chip enables anal- 94. Vatine, G.D. et al. (2019) Human iPSC-derived blood-brain
ysis of human lung inflammation and drug responses in vitro. barrier chips enable disease modeling and personalized
Nat. Methods 13, 151–157 medicine applications. Cell Stem Cell 24, 995–1005
70. Binnewies, M. et al. (2018) Understanding the tumor immune 95. Millard, D.C. et al. (2016) Identification of drug-drug interac-
microenvironment (TIME) for effective therapy. Nat. Med. 24, tions in vitro: a case study evaluating the effects of sofosbuvir
541–550 and amiodarone on hiPSC-derived cardiomyocytes. Toxicol.
71. Ando, Y. et al. (2019) Evaluating CAR-T cell therapy in a Sci. 154, 174–182
hypoxic 3D tumor model. Adv. Healthc. Mater. 8, e1900001 96. Ellis, B.W. et al. (2017) Human iPSC-derived myocardium-on-
72. van den Berg, A. et al. (2019) Personalised organs-on-chips: chip with capillary-like flow for personalized medicine.
functional testing for precision medicine. Lab Chip 19, Biomicrofluidics 11, 024105
198–205 97. Ulmer, B.M. et al. (2018) Contractile work contributes to matu-
73. Jenkins, R.W. et al. (2018) Ex vivo profiling of PD-1 blockade ration of energy metabolism in hiPSC-derived cardiomyocytes.
using organotypic tumor spheroids. Cancer Discov. 8, 196–215 Stem Cell Reports 10, 834–847
74. Wang, X. et al. (2019) Dynamic profiling of antitumor activity of 98. Acun, A. and Zorlutuna, P. (2019) CRISPR/Cas9 edited
CAR T cells using micropatterned tumor arrays. Adv. Sci. induced pluripotent stem cell-based vascular tissues to
(Weinh) 6, 1901829 model aging and disease-dependent impairment. Tissue Eng.
75. Boussommier-Calleja, A. et al. (2016) Microfluidics: a new tool for A 25, 759–772
modeling cancer-immune interactions. Trends Cancer 2, 6–19 99. Acun, A. et al. (2019) In vitro aged, hiPSC-origin engineered
76. Polini, A. et al. (2019) Towards the development of human heart tissue models with age-dependent functional deteriora-
immune-system-on-a-chip platforms. Drug Discov. Today 24, tion to study myocardial infarction. Acta Biomater. 94,
517–525 372–391
77. Miller, C.P. et al. (2020) Engineering microphysiological 100. Goldfracht, I. et al. (2019) Engineered heart tissue models from
immune system responses on chips. Trends Biotechnol. 38, hiPSC-derived cardiomyocytes and cardiac ECM for disease
857–872 modeling and drug testing applications. Acta Biomater. 92,
78. Livingston, C.A. et al. (2016) Facilitating the commercialization and 145–159
use of organ platforms generated by the microphysiological sys- 101. Park, S.E. et al. (2019) Organoids-on-a-chip. Science 364,
tems (Tissue Chip) program through public-private partnerships. 960–965
Comput. Struct. Biotechnol. J. 14, 207–210 102. Takebe, T. et al. (2017) Synergistic engineering: organoids
79. Zhang, B. and Radisic, M. (2017) Organ-on-a-chip devices meet organs-on-a-chip. Cell Stem Cell 21, 297–300
advance to market. Lab Chip 17, 2395–2420 103. Homan, K.A. et al. (2019) Flow-enhanced vascularization and
80. Mastrangeli, M. et al. (2019) Organ-on-chip in development: maturation of kidney organoids in vitro. Nat. Methods 16,
towards a roadmap for organs-on-chip. ALTEX 36, 650–668 255–262
81. Lind, J.U. et al. (2017) Instrumented cardiac microphysiological 104. Achberger, K. et al. (2019) Merging organoid and organ-on-a-
devices via multimaterial three-dimensional printing. Nat. chip technology to generate complex multi-layer tissue models
Mater. 16, 303–308 in a human retina-on-a-chip platform. Elife 8, e46188
82. Zhu, J. et al. (2018) An integrated adipose-tissue-on-chip 105. Low, L.A. and Tagle, D.A. (2016) Tissue chips to aid drug
nanoplasmonic biosensing platform for investigating obesity- development and modeling for rare diseases. Expert Opin.
associated inflammation. Lab Chip 18, 3550–3560 Orphan Drugs 4, 1113–1121
83. Liu, Y. et al. (2019) High-spatial-resolution multi-omics atlas 106. de Mello, C.P.P. et al. (2019) A human-on-a-chip approach to
sequencing of mouse embryos via deterministic barcoding in tackling rare diseases. Drug Discov. Today 24, 2139–2151
tissue. bioRxiv Published online October 1, 2020. https://doi. 107. Kaitin, K.I. (2010) Deconstructing the drug development
org/10.1101/788992 process: the new face of innovation. Clin. Pharmacol. Ther.
84. Zhang, Y.S. et al. (2016) Bioprinting 3D microfibrous scaffolds 87, 356–361
for engineering endothelialized myocardium and heart-on-a- 108. Blumenrath, S.H. et al. (2020) Tackling rare diseases: clinical
chip. Biomaterials 110, 45–59 trials on chips. Exp. Biol. Med. 245, 1155–1162
85. Li, J. et al. (2020) Improving bioprinted volumetric tumor micro- 109. Flecknell, P. (2002) Replacement, reduction and refinement.
environments in vitro. Trends Cancer 6, 745–756 ALTEX 19, 73–78

Trends in Pharmacological Sciences, February 2021, Vol. 42, No. 2 133