One of the most challenging aspects of drug development is testing. Scientists are forced either to experiment on whole animals, which is expensive, raises ethical issues and may not predict effects in humans, or to perform tests on microscopic human cells found in tissue cultures, which have been altered to live forever and bear little relation to actual living, breathing people. But researchers are working on a new technique to help bridge that gap: microchips that simulate the activities and mechanics of entire organs and organ systems. These “organs on a chip,” as they are called, are typically glass slides coated with human cells that have been configured to mimic a particular tissue or interface between tissues. Developers hope they could bring drugs to market more quickly and, in some circumstances, perhaps even eliminate the need for animal testing.

The chips are still in their early stages, but investigators are translating more and more body parts to the interface. Last summer bioengineers at Harvard University wrote in the journal Science that they had created a device that mimics a human lung: a porous membrane surrounded by human lung tissue cells, which breathes, distributes nutrients to cells and initiates immune responses. In November 2010 Japanese researchers announced online in An­­alytical Chemistry that they had built a chip that simultaneously tests how liver, intestine and breast cancer cells respond to cancer drugs, and in February 2010 scientists publishing in the Proceedings of the National Academy of Sciences USA developed a microscale replica of the human liver that allowed them to observe the entire life cycle of hepatitis C, a virus that is difficult to observe in cultured cells.

Pharmaceutical companies have expressed interest in the chips but are proceeding with caution. The main drawback, some say, is that the chips may not capture certain crucial aspects of living physiology the way whole animal tests do. “If you don’t use as close to the total physiological system that you can, you’re likely to run into troubles,” like being surprised by side effects later on in clinical trials, says William Haseltine, founder and former chairman and CEO of Rockville, Md.–based Human Genome Sciences. Harvard researchers say the chips can provide hints about toxicity: for instance, the lung-on-a-chip initiated an immune response against silica nanoparticles, which are under investigation as possible drug-delivery vehicles.

Ultimately, the goal is to make chips that mimic more complex systems—perhaps even entire humans, says Donald Ingber, director of Harvard’s Wyss Institute for Biologically Inspired Engineering and co-creator of the lung-on-a-chip. Scientists could build chips contain­ing cells from patients with specific genetic mutations, which could predict drug responses in specific populations, as well as personalized chips that predict an individual’s drug response. “Essentially this would be analogous to human clinical trial design, but all on inexpensive chips,” Ingber says. “This is the whole point of bioinspired engineering. You don’t have to re-create everything—you just have to get the salient features in.”