An organ on a chip is a microfluidic device, roughly the size of a USB stick, that mimics the structure and function of a living human organ. The core idea is surprisingly literal: living human cells are seeded into tiny channels etched into a transparent material, then exposed to fluid flow and mechanical forces that replicate what those cells experience inside the body. The technology emerged from a landmark 2010 demonstration of a lung on a chip and has since expanded to model the heart, liver, gut, brain, kidneys, and more. It sits at the intersection of cell biology, microfabrication, and pharmacology, and its most immediate promise is giving drug developers a way to predict human responses without relying so heavily on animal testing.
How the Technology Works
At its simplest, an organ chip is a small block of flexible, transparent material with hollow channels running through it. Different cell types line different channels, separated by a thin porous membrane. Fluid is pumped through the channels to simulate blood flow or air movement, and the flexible material can be stretched rhythmically to mimic breathing motions or heartbeats. The original lung chip, for example, placed human lung airway cells on one side of a membrane and blood vessel cells on the other, then applied cyclic vacuum-driven stretching to recreate the mechanical strain of breathing. That device reproduced the lung’s response to bacteria and inflammatory signals at an organ level, not just a cellular one.
1PubMed Central. Reconstituting organ-level lung functions on a chipThe mechanical environment matters more than you might expect. Cells do not just sit passively; they sense and respond to the physical forces around them. Blood vessel cells, for instance, align and orient themselves depending on how fluid flows across their surfaces. In microfluidic devices that generate controlled flow patterns, endothelial cells polarize in response to shear stress and orient upstream when that stress is strong enough.
2PubMed Central. Perpendicular alignment of lymphatic endothelial cells in response to spatial gradients in wall shear stress Getting these forces right is what separates organ chips from a standard dish of cells growing in a lab. Cells in a flat dish lack the architecture, flow, and stretch they need to behave like they would in your body. On a chip, they form tighter barriers, produce realistic levels of proteins, and metabolize drugs more like actual tissue does.
Most organ chips to date have been built from a silicone-based material called PDMS, which is easy to mold and optically clear, making it great for imaging living cells. The catch is that PDMS absorbs small drug molecules, which creates a serious problem when the whole point is to study how cells respond to precise drug concentrations. The field is gradually moving toward alternative materials that do not soak up drugs and are more compatible with mass production.
3ACS Biomaterials Science & Engineering. Beyond Polydimethylsiloxane: Alternative Materials for Fabrication of Organ-on-a-Chip Devices and Microphysiological SystemsWhich Organs Have Been Built on Chips
The catalog of organ chips has grown rapidly. Each model is designed to capture the specific features that matter most for that organ’s function and its relevance to disease or drug testing.
The liver chip is arguably the most commercially important, because drug-induced liver injury is one of the top reasons medications fail in clinical trials or get pulled from the market after approval. Liver chips seed primary liver cells alongside the specialized endothelial and immune cells found in the liver’s blood-filtering sinusoids. One advanced model demonstrated the ability to metabolize anticancer drugs and measure the downstream effects on liver-specific functions like enzyme activity and albumin production.
4PubMed Central. Advanced Liver-on-a-Chip Model for Evaluating Drug Metabolism and Hepatotoxicity A multispecies version of the liver chip has also been developed using rat, dog, and human hepatocytes, allowing researchers to compare toxicity across the same species used in preclinical animal studies and flag cases where an animal finding may not translate to humans.5PubMed. Reproducing human and cross-species drug toxicities using a Liver-Chip
The gut chip tackles a different challenge: recreating the dynamic relationship between human intestinal cells and the trillions of microbes living alongside them. A human gut-on-a-chip successfully co-cultured multiple species of commensal gut bacteria in direct contact with living intestinal cells for over a week, something nearly impossible in a standard dish because the bacteria tend to overgrow and kill the human cells. The device allowed researchers to tease apart how the microbiome, immune cells, and the rhythmic squeezing of peristalsis each contribute to bacterial overgrowth and inflammation.
6PubMed Central. Contributions of microbiome and mechanical deformation to intestinal bacterial overgrowth and inflammation in a human gut-on-a-chipHeart chips use cardiomyocytes, often derived from stem cells, cultured under mechanical stretch and fluid flow to produce beating tissue. A multi-lineage heart chip demonstrated the ability to detect the toxic effects of sorafenib, a cancer drug known to cause heart damage in patients. The chip picked up reduced contraction strength at low drug concentrations and disrupted calcium signaling at higher ones, closely mimicking what clinicians see in real cardiotoxicity.
7PubMed Central. Multi-Lineage Heart-Chip Models Drug Cardiotoxicity and Enhances Maturation of Human Stem Cell-Derived Cardiovascular CellsThe blood-brain barrier is one of the trickiest structures to model because it is so selective, blocking nearly everything from crossing from the bloodstream into the brain. A perfused blood-brain barrier chip formed tight junctions between cells and severely restricted the passage of large molecules, closely mimicking the real barrier’s selectivity.
8PubMed Central. A perfused human blood-brain barrier on-a-chip for high-throughput assessment of barrier function and antibody transport This is especially relevant for developing drugs that target neurological diseases, where getting a therapeutic molecule past the barrier and into the brain is often the hardest engineering problem.
Connecting Multiple Organs
A single organ chip can tell you a lot about one tissue’s response, but drugs do not affect just one organ. Your liver metabolizes a drug into breakdown products that may be harmless to the liver but toxic to the heart. To capture those systemic interactions, researchers have begun linking multiple organ chips together through shared fluid channels, creating what is sometimes called a “body on a chip.” These linked systems draw on pharmacokinetic modeling principles to set the relative sizes and flow rates between organ compartments so they approximate real physiological ratios.
9PubMed Central. Using physiologically-based pharmacokinetic-guided “body-on-a-chip” systems to predict mammalian response to drug and chemical exposureA pumpless heart-and-liver chip demonstrated this idea concretely using the chemotherapy drug doxorubicin. The liver compartment metabolized doxorubicin into doxorubicinol, a known cardiotoxic byproduct. Heart cells connected to the liver compartment showed significantly more damage than heart cells exposed to doxorubicin alone in a static dish, precisely because they were encountering both the parent drug and its metabolite, just as a patient’s heart would.
10Organs-on-a-Chip. Heart/liver-on-a-chip as a model for the evaluation of cardiotoxicity induced by chemotherapies Linking organ chips also opens the door to modeling how a drug absorbed through a gut chip is processed by a liver chip before reaching a heart or brain chip downstream.11PubMed. Physiologically Based Pharmacokinetic and Pharmacodynamic Analysis Enabled by Microfluidically Linked Organs-on-Chips
Personalized Medicine Through Patient-Derived Cells
One of the most exciting frontiers is seeding organ chips with cells derived from a specific patient. Using induced pluripotent stem cell technology, researchers can take a blood sample or skin biopsy, reprogram those cells into a stem-cell-like state, then differentiate them into the cell types needed for a chip. The result is a miniature organ that carries the patient’s own genetic makeup.
12PubMed. iPSC-derived organ-on-a-chip models for personalized human genetics and pharmacogenomics studiesThis is not purely theoretical. A blood-brain barrier chip built with patient-derived cells from individuals with neurological diseases revealed disease-specific deficits in drug transporters and disrupted barrier integrity that would never show up in a chip made from generic healthy cells.
13Cell Stem Cell. Human Brain Microvascular Endothelial Cell Human Blood-Brain Barrier Organ-Chip Exposes Interindividual Variability in Drug Permeability and Transport The implication is that you could, in principle, test a drug on a chip built from your own cells before your doctor prescribes it, identifying whether you are likely to respond well or experience side effects. That is still a long way from routine clinical use, but the proof of concept already exists.
Cancer Research and Metastasis Modeling
Cancer chips tackle a problem that is notoriously hard to study in animals or in standard cell cultures: how cancer spreads. Metastasis involves tumor cells entering the bloodstream, traveling to a distant site, and squeezing through blood vessel walls to establish a new tumor. Microfluidic models can reconstitute the blood vessel lining and the surrounding tissue, then introduce circulating tumor cells to watch, in real time, how they arrest and push through the vessel wall.
14PubMed Central. Microfluidic vascular models of tumor cell extravasationThese cancer-on-chip models have been used to study breast cancer cell extravasation into bone tissue, track how quickly tumor cells cross the endothelial barrier, and distinguish between partial protrusion and full transmigration. The ability to quantify each step of that process and test how different drugs or microenvironmental factors alter it is something animal models cannot offer with the same clarity.
15Trends in Biotechnology. Cancer-on-chip models for metastatic disseminationModeling Infectious Disease
The COVID-19 pandemic demonstrated both the need for and the capability of organ chips in infectious disease research. A human alveolar chip was used to model SARS-CoV-2 infection by co-culturing lung epithelial cells, blood vessel cells, and circulating immune cells under fluid flow. When the virus was introduced, the chip reproduced key features of severe COVID lung injury: immune cells were recruited to the infection site, endothelial cells detached from the vessel wall, and inflammatory signaling molecules surged.
16PubMed Central. Biomimetic Human Disease Model of SARS-CoV-2-Induced Lung Injury and Immune Responses on Organ Chip System The presence of circulating immune cells turned out to be crucial for reproducing the exaggerated inflammatory response that drives much of the damage in severe cases, reinforcing that immune cell integration is not optional for realistic disease modeling.
Building immune function into organ chips is an active area of development in its own right. Practical guides now exist for designing chips that model how immune cells are recruited from the bloodstream into infected tissue during acute infections.
17PubMed Central. Integration of immune cells in organs-on-chips: a tutorial Other models have recreated inflammatory responses in coronary artery chips by adding circulating monocytes that attach to and invade vessel walls when inflammation is triggered, mimicking early stages of atherosclerosis or vascular damage.18Interdisciplinary Medicine. Human coronary artery organ‐chip with circulating immune cells recapitulates anti‐inflammatory effect of pulsatile wall strain
Built-In Sensors and Real-Time Monitoring
Early organ chips required researchers to remove samples and analyze them off-chip, which disrupted the living system and gave only snapshot data. Newer designs integrate sensors directly into the device. One recent system combined sensors for oxygen concentration, pH, and electrical resistance (a measure of barrier tightness) on a single chip, mounted on a printed circuit board and run using a commercial instrument. It monitored tissue responses continuously and noninvasively for up to 23 days, tracking how cells reacted to chemical stimuli in real time.
19PubMed Central. Organ chips with integrated multifunctional sensors enable continuous metabolic monitoring at controlled oxygen levelsLiver chips have similarly incorporated real-time electrical and optical sensors to monitor barrier integrity and pH changes continuously, providing ongoing readouts of drug-induced liver injury as it develops rather than only catching the damage after the fact.
20Journal of Micromechanics and Microengineering. Real-time physiological sensor-based liver-on-chip device for monitoring drug toxicity This shift from endpoint measurements to continuous monitoring is a significant upgrade. It means researchers can watch a toxic response unfold over hours or days, pinpointing exactly when damage begins and whether it is reversible.
Vascularization and Growing Organoids on Chips
One persistent limitation of three-dimensional tissue models like organoids (self-organizing miniature organ structures grown from stem cells) is that they lack blood vessels. Without perfusion, nutrients and oxygen cannot reach the interior of larger structures, limiting their size and maturity. A microfluidic platform has now demonstrated a way to grow functional vascular networks on-chip that form working connections with organoids, providing intravascular perfusion to these structures. Organoids cultured with this vascularization method showed enhanced growth, maturation, and function compared to those grown without it.
21Nature Communications. A microfluidic platform integrating functional vascularized organoids-on-chip Merging organoid biology with organ-chip engineering is increasingly seen as the path toward more complex and realistic tissue models.
The Regulatory Shift Away from Mandatory Animal Testing
For decades, regulators required animal testing before any new drug could enter human clinical trials. That changed with the FDA Modernization Act 2.0, signed into law in the United States in late 2022, which removed the federal requirement that drugs be tested on animals before human trials. The law explicitly allows the use of “new approach methodologies,” a category that includes organ-on-a-chip systems and patient-derived organoids. The FDA’s 2025 roadmap is establishing frameworks for integrating these technologies into the drug development pipeline.
22PubMed Central. Human-Centric Modeling in Metastatic Breast Cancer: Organoids, Organ-on-Chip Systems, and New Approach Methodologies in the Post-FDA Modernization Act 2.0 EraThis does not mean animal testing has disappeared overnight. The pharmaceutical industry moves slowly, and regulators still want to see robust validation that chip-based results predict human outcomes reliably. But the legal barrier is gone, and the economic incentive is powerful: drug development is extraordinarily expensive, with high failure rates in late-stage clinical trials often driven by toxicity that animal models failed to predict. High-throughput organ-chip systems are seen by pharmaceutical companies as a potential way to cut those costs by filtering out toxic candidates earlier.
23PubMed Central. State-of-the-art in high throughput organ-on-chip for biotechnology and pharmaceuticalsOrgan Chips in Space
An unexpected application has emerged in space medicine. Microgravity accelerates many aspects of aging and disease in ways that are difficult to study on Earth, and sending organ chips to the International Space Station offers a controlled way to isolate gravitational effects on human tissue. An automated heart-on-a-chip platform flew to the ISS for a month-long mission, with contractile function monitored in real time throughout. When the heart tissues returned to Earth, they showed significantly reduced contraction force, increased arrhythmias, and signs of structural damage to both the muscle fibers and mitochondria. Gene analysis revealed upregulation of pathways linked to heart failure, oxidative stress, and inflammation.
24PubMed Central. Spaceflight-induced contractile and mitochondrial dysfunction in an automated heart-on-a-chip platformA separate device integrated organ-chip technology with a clinostat (a device that simulates microgravity on the ground) to study how altered gravity affects blood vessels. Simulated microgravity induced signs of premature aging in the endothelial and smooth muscle cells lining the artificial vessels.
25PubMed. A biosensory μvessel-gravity device for advancing vascular analysis in space medicine These space studies are not just about astronaut health; the accelerated aging seen in microgravity may help researchers understand age-related vascular disease on Earth in compressed timeframes.
Ethical Questions Around Patient Cells and Consent
As organ chips increasingly rely on patient-derived cells, ethical questions have surfaced that are not fully resolved by existing consent frameworks. Traditional biobanking consent asks a donor to agree, at one point in time, to broadly defined future research uses of their tissue. The problem is that organ-chip technology enables uses that a donor may not have imagined when they signed a consent form. A person’s cells might be used to build a chip that tests a controversial drug, models a stigmatized disease, or gets incorporated into a commercial product. Scholars have argued that standard consent processes fall short here because a donor may value their biological materials differently depending on the specific context of use.
26PubMed Central. Organoid biobanking, autonomy and the limits of consentThe conversation extends further as the technology evolves. As chips grow more complex and potentially incorporate cells from multiple donors, or as “clone-on-a-chip” concepts emerge that could model an individual’s full biology, the ethical landscape expands to include questions about property rights, commercialization, and societal implications that go well beyond what informed consent was originally designed to address.
27BMEMat. Toward clone‐on‐a‐chip: Ethical and legal considerations for organ‐on‐a‐chip evolutionScaling Up and Making It Practical
For all the scientific promise, the field faces real engineering and manufacturing hurdles. Most organ chips are still handcrafted in academic labs, one at a time, by trained researchers. Moving to industrial-scale production requires overcoming the PDMS absorption problem mentioned earlier, developing standardized designs that different labs and companies can reproduce, and integrating chips with automated handling systems so they can be run at the throughput pharmaceutical screening demands. Several commercial companies are now selling organ-chip platforms, and the development of high-throughput systems specifically designed for drug discovery is an active area of industry investment.
28PubMed Central. State-of-the-art in high throughput organ-on-chip for biotechnology and pharmaceuticalsSustainability is another dimension that does not get as much attention. A review examining whether organ chips could meaningfully reduce the environmental footprint of drug testing raised the question of whether scaling up production of plastic-based microfluidic devices, cell culture media, and associated consumables might create its own waste problem.
29PubMed. Organ-on-a-Chip for Drug Screening: A Bright Future for Sustainability? A Critical Review The field’s sustainability case rests partly on replacing large numbers of laboratory animals and their associated resource costs, but whether the net environmental impact is favorable depends on how manufacturing scales. It is a question the field is just beginning to grapple with, and one that will matter more as production volumes grow.