10 min read

Organ-on-a-chip: Microfluidics and drug discovery

Abstract: Drug discovery is a long, tedious and expensive process with multiple rounds of tests. A big bottleneck in the process is that drugs are first tested on animals first which don't translate really well to humans. 9 out of 10 drugs that prove successful in animal models fail in human tests i.e 2-5 billion$ of cost and 10-12 years of time lost in the drug discovery process. The organ-on-a-chip approach aims to mimic the biological process in-vitro and thereby reduce both time as well as costs associated with the drug discovery process.


One of the greatest manifestations of human ingenuity is the quest to solve problems. Ever since primitive times, we have sought reasons for existing problems, cultural and biological, and tried to fix them. At the biological level, 'treatments' for common conditions have been part of different societies that were passed through ages. Humans since ancient days have searched for elixirs or natural medicines that could destroy diseases in our bodies and keep us healthy. And our ancestors did discover certain effective natural medicines.

When humans began using formal writing systems, they also began documenting their use of medicinal herbs around the world, so we can be more certain about the history of herbal remedies dating back to about  3000 B.C.  Illustration by Sophie Kittredge. Source: Motherearthliving.com

Perhaps, the discovery of certain natural medicines came during the search for edible plants. There is no evidence to prove that but humans have certainly experimented with many things. Most experiments with herbs and plants as forms of natural remedies have failed while the rest either became helpful drugs(medication and recreation) or harmful drugs. Interestingly, the origin of the word 'drug' in Dutch and French languages takes root from the barrels once used to keep herbs dry. Although, one could say that any such 'drug' before the scientific age was muddled between superstitious and religious beliefs. Sages or religious leaders were often the administrators of drugs.

The first big breakthrough came only around 500 BC when Hippocrates had an insight into the fact that diseases could have natural rather than supernatural causes. The great Chinese, Indian, Arabic, Roman and Greek Civilizations of the past discovered drugs from herbs, and sometimes even from animals and minerals. However, drug discovery and development started to follow scientific techniques starting only in the late 1800s.

The discovery of penicillin in 1928 by Alexander Flemming foreshadowed the commencement of drug developments driven by biotechnology, where microorganisms were harnessed. The development of recombinant DNA products that utilized cellular and molecular biology became more prominent in the late 1970s. This began the era of pharmaceutical industries.

Pharmaceutical drugs have generally benefited human health as well as increased human longevity, but there have been quite a few monumental challenges. While the influence of profit-driven corporations(Valeant Pharmaceuticals) have driven the prices of certain drugs, the high cost of discovering a new drug can be alluded to how R&D works in the pharmaceutical industry. It takes around 10 years while roughly 80% of the drug candidates fail in clinical trials while the costs could soar up to a couple of billion dollars.

Costs are in Million $ on the y-axis

Drug discovery process:

Drug Discovery Process. Credits: www.nebiolab.com

Drug discovery is a long, tedious and expensive process with multiple rounds of tests. Pre-clinical tests usually are made on animals such as mice or cell culture. Clinical trials are then done on humans if a drug clears the pre-clinical stage.

Why are drug discoveries so expensive?

Costs differ for each drug, depending on factors such as the number of iterations during the discovery and pre-clinical phases, the experimental design(s) required during pre-clinical and clinical trials, and more. However, the major reason why drug discoveries are so expensive on average can be alluded to:

a. High failure rate: While a large share of expenses come at the clinical stage, almost 70% of the drugs fail at the pre-clinical stage while taking 50% of the time in the entire process. One key factor that plays a huge role here is that animals are used to test the drug at the pre-clinical stage. Besides the fact there is heated debate surrounding the ethics of using animals in medical research, animal models are not predictive of the human situation. Perhaps, even hindering the progress in drug discovery. 9 out of 10 drugs that prove successful in animal models fail in human tests i.e 2-5 billion$ of cost and 10-12 years of time lost in the drug discovery process. Decades of attempts to develop treatments for diseases including asthma, cancer, stroke, and Alzheimer's using animals have failed to translate to humans.

b. The need for personalization: In the US there are now over 7,000 conditions recognised as rare diseases. It is estimated almost nine of ten of these conditions currently have no approved treatment. These are the diseases that need personalized, precision medicines. Each individual reacts differently.

c. Regulatory Bottlenecks: While it's already a complex and rigorous scientific process to create a medicine that we can be sure is both effective and safe, FDA and other regulatory bodies have placed stringent regulations in line with protection over experimentation. Regulatory bodies should consider re-framing their approach.

There are currently two major approaches in solving the challenges head-on at pre-clinical stages:

  1. Applying Machine learning (ML) tools in the drug discovery process improving discovery and decision making. Examples include target validation, identification of prognostic biomarkers and analysis of digital pathology data in clinical trials.
  2. Organ-On-a-Chip (OOAC) is a biomimetic system that can mimic the environment of a physiological organ. The major goal of OOAC is to simulate the physiological environment of human organs. OOAC devices could provide insights into normal human organ function and disease pathology, while predicting the safety and efficacy of drugs in humans more accurately.

Let’s dive into OOAC today.

Credits: ufluidix.com

Technical Landscape

The reason why animal models don't translate to humans is that the organ functions and tissue interactions are not the same between humans and animals. If drugs can instead be tested on a realistic human OOAC model, harmful or ineffective drugs can be eliminated much earlier in the drug discovery process. Thereby, reducing failure while saving money and time. By combining cell biology, engineering and biomaterial technology, OOAC chip provides a micro-environment to simulate an organ in terms of tissue chips and mechanical simulation.

So how does it work? OOAC chips are essentially living, 3D structures incorporating major functional units and complexities of a living organ. The in-silico chip is the size of a computer memory stick composed of a clear flexible polymer. The polymer contains hollow microfluidic channels - sub-millimetre channels - that can control the precise behaviour of fluids within a particular geometry. These channels are lined with organ-specific human cells interfaced with an artificial vasculature - a network of blood vessels. Mechanical forces are then applied to mimic a physical environment of an organ such as breathing motion in lungs or deformations in intestines.

Breakdown of OOAC components. Credits: Wyss Institute, Harvard

Wyss Institue, for example, has developed an airway-on-a-chip to understand the breathing function in the lungs. The chip contains two hollow channels with a porous membrane(air chip) in between. Living cells are aligned on top of the membrane and capillary cells(blood vessels that deliver oxygen and collect carbon dioxide) on the bottom side like an actual lung. While air flows above the living cells, white blood cells flow below the capillary cells and the mucus flows through the air chip.

The whole process reciprocates an actual lung. The white blood cells are then stimulated to inflammatory signals just as in a human body, providing a profile of different interactions among the constituents. This disease modelling enables the identification of new lung disease biomarkers.

In general, the channels can used to deliver nutrients, bacteria, viruses, and chemicals to the cells or wash the cells and discard the waste. They can also be employed to manipulate the cells mechanically or electrically.

There are a variety of projects developing microchips that recapitulate the microarchitecture and functions of living human organs such as the lung, intestine, kidney, skin, bone marrow and blood-brain barrier. The long-term vision is to create a 'body-on-chip' model. By linking various organ-on-chip models, it would be possible to simulate the interaction between organs. This will enable studying the effect of a drug not only on a particular organ but also the impact of other organs on the drug or vice versa.

Credits: elveflow.com

Organ-on-a-chip can also greatly impact the personalized medicine space. By introducing a specific patient's cells in the model and pre-screening different drug candidates, the administration of the right drug for the specific patient can be ensured.

This technology has the potential to extend beyond the pharma industry into the cosmetics and food industry as well. Skin-on-chip to test the side -effects of cosmetics or intestine-on-chip to test allergies to a new food product. Another interesting application is personal diagnostics. By extracting cells from an individual and placing it in organ-on-a-chip, biomarkers that characterize diseases can be discovered.. Thereby, identifying the diseases much earlier.

The technology is still in early stages, device engineering and manufacturing remain as major bottlenecks hampering the launch of OOAC technology to the industry. There is a great need for further development and testing activities to reach a broad adoption and acceptance of OOAC by large pharmaceutical companies and regulatory agencies. While the technology is promising, a recent extensive survey claims that the technology readiness level(TRL) of OOAC is TRL 4 (on a scale of 10). A particular technology is considered ready for commercial applications only at TRL 9.

There are many bottlenecks that need to be overcome before the technology hits the market. Some of them are listed below:

  • We discussed earlier about human cells being lined inside the chip. However, the consumption, metabolic conversion, and secretion of energy and respiration compounds vary based on:

    • cell sources (i.e., immortalized cell lines,primary cells from donor tissue, human induced pluripotent stem cells)

    • inter-laboratory storage and handling conditions

    • nutrient availability

    • tissue maturity

    • developmental stage(e.g., embryonic development, pregnancy)

    • genetics

    • disease status, etc.

  • Another big reason the industry is still nascent can allude to a lack of standardization in device design or fabrication material. Standardization would make it easier to cross-lab validate different OOAC models, simplify the integration of OOAC models with existing equipment and potentially increase the adoption rate of the technology.

  • Translation of OOAC models to predictive clinical outcomes remains a challenge. For example, certain design trade-offs provide testing capacity only for a short period of time making most models unsuitable for chronic exposure which needs to be studied over long durations.

  • Other key issues raised by both developers and end-users include device robustness, reproducibility, affordability, and ease of use. Finally, the feasibility of scaling body-on-a-chip platforms at industrial levels will be a very hard challenge to overcome as there will be a need for manual (unnatural/artificial) labour involved in integrating different organ(s)-on-chips, increasing both functional and structural complexity.

  • The human body is a complex system. Studying the impact of drugs on independent organs may not be the same as studying the impact on interconnected organs.

Despite the challenges and long wait before commercial validation, so many startups are working towards making OOAC technology a reality.

Market Landscape

Emulate is a US start-up, a spin-off from Wyss Institute for Biologically Inspired Engineering, released with the first organs-on-chips. The company has developed many organs-on-chip (such as lungs-on-chip, gut-on-chip and even blood-brain-barrier-on-chip), and are currently researching personalized medicine.

InSphero AG, a pioneer in 3D cell-based assay technology, that spun from the Bio Engineering Laboratory of ETH Zurich, today announced that they have been working on commercialization of their Akura Flow organ-on-a-chip platform to ensure operational robustness and trustworthy results demanded by the pharmaceutical industry. The platform aims to integrate 3D spheroid models in single- and multi-tissue organ networks for preclinical drug efficacy and toxicity testing applications.

Mesobiotech is a French company that manufactures organs-on-chip for personalized medicine. AxoSim researches nerves-on-chip to use the special microfluidic chips to fight cancer. Tara Biosystems is a US company that develops heart-on-chip to produce a heartbeat-like movement by exposing the chip to electrical stimuli. Bi/ond is a Dutch company developing OOAC models to recreate aspects of human organs such as the beating of a heart, tissue interaction and blood circulation. The company aims to recreate human physiology and pathology through OOAC technology to enable the era of personalized precision medicines.

Nortis Bio, another US startup develops kidneys-on-chips and other organs-on-chips to advance research on new treatments and reduce both cost and time of development.  AlveoliX is a Swiss startup that has developed a human lung-on-a-chip model. By mimicking the biophysical microenvironment of the air-blood barrier of the human lung, the OOAC model is expected to best predict the effects of respiratory drug candidates in humans.

TissUse, a German start-up spun from the Technische University in Berlin produces multi-organ-on-chip such as HUMMIC - 4 organs on the same chip - and is currently developing a new human-on-chip with 10 organs. BioMimix, an Italian startup, Spinoff of Politecnico di Milano is developing a Beating-Heart, Organ-on-a-Chip enabling one to apply different experimental setup in terms of cells and culture conditions, to speed up the discovery of new treatments for Big Diseases, towards a Precision Medicine future.

SpacePharma is another company focused on leveraging microgravity to probe potential major health issues in space using organ-on-a-chip technology.


Elveflow is a kick-ass institute developing a range of state-of-the-art microfluidic instruments, for all kinds of microfluidic applications and they have great resources on organ-on a chip.

European Organ-on-a-chip(EUROOCS) is a European society to build a bigger ecosystem around the technology by bringing in different stakeholders together to make further progress and commercialize the technology.

Comments powered by Talkyard.


Become a Infiniti Ventures member below to join the conversation (it's free!). As a member, you will also receive new posts by email (you can unsubscribe at any time).