Images courtesy of Dino Di Carlo. Artwork by Marc Lim.
"Drug toxicity is one of the most common reasons why promising compounds fail. We need to know which ones are safe and effective much earlier on in the process. This is an unprecedented opportunity to speed development of effective therapies, while saving time and money." – Francis S. Collins, M.D., Ph.D., director, National Institutes of Health (NIH News, September 16, 2011)
In September 2011, the White House invited proposals to bring together the latest advances in engineering, biology and toxicology to create a chip (a microengineered device) that can be used to screen for safe and effective drugs faster and more efficiently than current methods, and before they are tested on humans. According to NIH, the vision for this chip is that it will be loaded with specific cell types that reflect human biology, and will be designed to allow multiple different readouts that can indicate toxicity. As much as $70 million will be invested in the effort.
"We know the development pipeline has bottlenecks in it, and everyone would benefit from fixing them," Collins said. "What we need are entirely novel approaches to translational science, to take full advantage of the deluge of new biomedical discoveries that have been made in recent years."
The September 16, 2011 issue of ScienceInsider reports, "The drug chip will be developed in a partnership to include NIH, the Food and Drug Administration (FDA), and DARPA – the Defense Advanced Research Projects Agency – which is known for funding risky research. Collins says the first-ever collaboration of this kind will try to combine human cell types, such as liver and kidney cells, that can represent physiological systems and ‘talk to each other' on a chip that will be used to predict whether a drug will be safe. Researchers will try to grow cells in three dimensions rather than as a flat layer because that's a better way to model how a drug will act in human tissues. The project "is really ambitious," Collins says.
SLAS member Dino Di Carlo, Ph.D., agrees. "I think it's definitely a challenge. Researchers have successfully grown different types of cells on chips – liver, intestinal and lung epithelial cells, and endothelial cells that form blood vessels – but usually no more than one or two cell types on the same chip. Just getting one cell type to grow is difficult over a long time period. There are contamination issues with bacteria or fungi; and growing cells can obstruct the channels used to provide nutrients.
"To put cells in on-chip environments to recapitulate organ systems with the complexity to test drugs instead of using animal models will be exponentially more difficult," says Di Carlo. "Media formulations for one cell type might stop growth, de-differentiate or outright kill another cell type. Errors will multiply. If you have a 90 percent success rate with one type of cell, your success rate will drop to 30 percent for 10 cell types just considering independent error in preparing the systems. Significant automation can be a big help here."
Di Carlo, a former SLAS Innovation Award finalist and current member of the SLAS Journal of Laboratory Automation Editorial Board recently was named a 2011 Packard Fellow for Science and Engineering, and was honored with a 2011 DARPA Young Faculty Award and an NIH Director's New Innovator Award and for his work in the Department of Bioengineering at the University of California Los Angeles. At UCLA, Di Carlo and his team investigate novel ways to interrogate and manipulate cells and automate their analysis. Di Carlo also has been taking advantage of the ability to control local environments in microscale chips to make more in vivo-like conditions. "We grow cancer cells in microenvironments that can be better controlled – both in terms of the chemical and mechanical environment – using microfluidic chips. We aim to understand how the environment affects cancer progression and to simulate a more in vivo-like situation to create realistic in vitro models for anti-cancer drug testing.
"This new NIH DARPA challenge is to combine microfabrication techniques with modern tissue engineering to allow different populations of cells to connect, communicate and form a more complete physiological system that can better mimic the behavior of human organs," says Di Carlo. "These on-chip systems would allow researchers to see secondary and more complicated effects in engineered organ systems when exposed to stimuli like drugs – effects they can't see with single cell models – that will help predict drug efficacy and toxicity."
DARPA's portion of the project focuses on engineering; abstracts were due in October and proposals were due in December 2011. The DARPA Microphysiological Systems project is officially summarized as follows: DARPA is soliciting innovative research proposals to develop an in vitro platform of human tissue constructs that accurately predicts the safety, efficacy, and pharmacokinetics of drug/vaccine candidates prior to their first use in man. Alternative testing methods that rely on isolated human cells hold the promise of authentic human responses to candidate drugs, vaccines, and biologics. Recent research has shown that three-dimensional constructs of one or more cell types are able to reproduce relatively authentic human tissue and organ physiology in an in vitro environment. As a result, DARPA seeks in vitro platforms comprised of human tissue constructs that will accurately assess efficacy, toxicity, and pharmacokinetics in a way that is relevant to humans and suitable for regulatory review.
An optimal outcome of this challenge is the avoidance of animal models in drug toxicity testing. "Animal models don't always reflect human responses, and in addition to being unreliable, they're expensive," says Di Carlo. "An ideal on-chip system using human cells would be a much, much better model for disease physiology. Drug discovery already uses cell cultures to create environments for drug testing. Instead of cells growing in plates, however, the cells growing on these chips would be in more three-dimensional structures and would include vasculature that allows chemical communication between cell populations. Researchers could find appropriate dosing in a realistic way."
But is this goal within reach? Di Carlo thinks so. "Engineered micro- and nanofluidic devices appear ideally suited to diagnose, simulate and probe biological systems. Some working pieces are already in place like the in vitro liver cell cultures." But knowing that sometimes pharmaceutical companies can be somewhat risk-averse, new technology likely will take time not only to develop, but to enter the mainstream. With this in mind, Di Carlo doesn't think we'll be seeing or using complete humans-on-a-chip anytime soon.
The existing working pieces mentioned by Di Carlo include efforts by several companies that have successfully transitioned from academic labs to produce and market chips or substrates that grow primary cells in ways that have functionality for pharma testing and validation. Examples of these companies include CellASIC. According to its website, "CellASIC was started out of the Bioengineering Department at the University of California, Berkeley. In early 2003, two graduate students, Paul Hung and Philip Lee, began work in the laboratory of Prof. Luke Lee to utilize microfabrication methods to improve laboratory cell culture methods. What began as a quest to engineer a ‘better Petri Dish' eventually led to the formation of CellASIC Corp." Today, CellASIC markets "precise microfabricated environments to improve the functionality of cell-based experimentation."
Another example is the HμREL Corporation, which markets "microfluidic devices and cell cultures that elevate the metabolic competency and endurance of cells cultured in vitro, and that simulate pharmacokinetic interactions among multiple tissues and organs," says its website. HμREL maintains worldwide exclusive rights to certain patents owned by Cornell University, where HμREL's microfluidic technology was invented by Michael Shuler and Gregory Baxter.
Hepragin, according to its website, "is developing a unique, bioengineered microliver platform (the HepatoPac Platform) as a highly functional model of the liver in vivo. Microfabrication and tissue engineering technologies have been combined to create a precise, organized and miniaturized human liver model. Micropatterned industry-standard multiwell plates contain tiny colonies of organized liver cells surrounded by supportive stroma." According to Sangeeta Bhatia, chair of the Hepragin Scientific Advisory Board, "My hope is that better in vitro models of liver tissue will translate into highly effective and affordable drugs which are the safest treatments for patients…"
At the Hansjorg Wyss Institute for Biologically Inspired Engineering at Harvard University, Founding Director Don Ingber and his Biomimetic Microsystems team are already working on trying to integrate their organs-on-a-chip, which include lungs, bone marrow and peristaltic gut, into predictive and practical microsystems. In an informative video on the Wyss Institute's website, Ingber reports they have experienced "great interest from pharmaceutical industry and hope to announce our first alliance to use those to see if we can indeed replace some animal studies to accelerate the drug development process and decrease costs."
"Where these new technologies might take us is exciting to think about," says Di Carlo. "There are many great minds hard at work on pushing the envelope – as the old saying goes, ‘genius is one percent inspiration, ninety-nine percent perspiration.' Major funding from sources like the NIH and DARPA really help pick up the pace. The current economy would welcome these advances as one way to directly address the high cost of healthcare. Plus, it's translational with easily envisioned job-creating industries. It could be a win-win all around."
December 19, 2011