One of the greatest challenges to our healthcare system today is creating effective new drugs. Despite ever-increasing investments in research and development, the number of drugs that win approval for clinical use each year has steadily decreased over the past 50 years. It now costs of more than $2.5bn (£1.6bn) to bring a single compound from the bench to patients. Because many have to be developed to find one that works, drug costs have become exorbitant.
Part of the problem is that the animal models we use to test potential therapeutics often fail to predict results in humans. Very simply, a guinea pig is not a human. To overcome this limitation, drug companies have tried to use human cell cultures to test drugs in the laboratory. But these too are poor predictors of how drugs will work in patients because they fail to accurately replicate the complexity of human organs. As a result, there has been a search for more accurate ways to mimic human organ functions outside the body.
We recognised part of the problem is that human cells are usually cultured on rigid plastic dishes in a static pool of medium, and in isolation from other cell types. In contrast, living organs are composed of two or more different types of tissues that interact with one another physically and chemically. These organs depend on flow of blood and other body fluids for their survival and function. They also experience dynamic physical motions, such as breathing in the lung and peristalsis (the movement of food using muscle contractions) in the intestine, which clinicians know to be crucial for normal physiology.
Knowing this, we distilled down the essence of organ design into three fundamental principles. First, the establishment of a tissue-tissue interface where organ-specific cells and blood vessel lining cells come together. Second, the provision of fluid flow to mimic blood flow. And third, reconstitution of physiological mechanical motions, such as rhythmic expansion and contraction of the air sacs in our lungs or peristalsis in our intestines.
Inspired by these key biodesign principles, we adapted manufacturing techniques from the computer microchip industry that allow the creation of cell-sized features to create artificial human “organs-on-chips”. These computer memory stick-sized devices are made of crystal-clear, flexible rubber and contain a central hollow channel. This channel is thinner than the width of a pencil lead and is separated into an upper and lower channel by a thin flexible porous membrane.
Lung on a chip
To make a lung-on-a-chip, for example, living human cells from the lung’s air sacs are cultured on the top of the membrane, while cells from human blood capillaries are placed on its lower surface. This replicates the normal tissue-tissue interface that mediates the gas exchange and the inflammatory response to infections in the lungs.
Air is then placed over the lung cells and a nutrient medium or whole blood containing human white blood cells is flowed through the lower channel. Finally, applying suction to the flexible tube’s side chambers causes the interfaced tissues to repeatedly stretch and relax. The result is a tiny replication of the process of breathing in and out that occurs in the lung.
When we introduced living bacteria into the air space of the lung chip, we triggered an infection within the chip. The white blood cells flowing through the vascular channel responded by migrating into the air space where they engulfed the invading germs, just as they do inside living lungs.
The lung chip also has been used to mimic diseases, such pulmonary edema or “fluid on the lungs”, and to identify new drug candidates that prevent this condition. Since then, we have fabricated many other organ chips, including a “gut-on-a-chip” that mimics the peristalsis movements that push food through the digestive tract. And our kidney chip faithfully mimics drug toxicities that occur in humans, but not in animal models. We have even started to link different types of organ chips by their channels to begin to create a human body-on-chips that can help analyse the way that drugs are absorbed, metabolised and cleared as they move throughout the body.
Given that it is now possible to create stem cells from normal cells taken from any individual, patient-specific organs-on-chips may one day be used to test how a drug might affect a particular person. However, the game-changer is how this technology could transform drug development. Most pharmaceutical companies carry out huge clinical trials that usually fail, and then they sift through the data searching for a subgroup of patients who might have responded better than others. They then initiate a new clinical trial on this small group of patients.