In a Restless Preclinical Testing Market, Organs-on-Chips Company MIMETAS Is in Expansion Mode
By Aaron Krol
October 31, 2014 | The CRACK IT Challenges, organized by the UK’s National Centre for the Replacement, Refinement & Reduction of Animals in Research (NC3Rs), have an ambitious goal: to foster projects that can replace animal studies as a basic test bed for biomedical research. Animal studies are part of the essential fabric of the life sciences industry, one of the first lines of defense against toxicity, off-target effects, and ineffective treatments. Yet there is a global movement afoot to diminish our reliance on animal testing, and while that’s partly driven by ethical concerns, more and more it’s being spearheaded by results-oriented pharmaceutical companies. While it is not new to observe that animal models are, at best, flawed representations of human biology, and at worst a dangerous and costly misdirection in drug development, what is new is a feeling that modern technology is finally putting better models within reach.
At the same time, key organizations pushing drug development away from the use of animals are careful to temper expectations for when and to what degree animal testing can realistically be replaced. It is telling that NC3Rs, one of the more well-resourced organizations in this movement, includes “refinement” of animal testing among its aims.
“We fully realize that animal testing is going to be here for decades at least,” says Jos Joore, the co-founder of MIMETAS, a Dutch company that develops alternative preclinical models for studying human disease. “There’s tons of legacy data, and there are regulatory bodies that have specific demands, so we really have to deal with the situation, as an industry, of how we can replace the bad models from the past.”
Nevertheless, when Joore and his colleagues at MIMETAS heard about a recent CRACK IT Challenge offering up to £1 million ($1.6 million) to build a new model of a human kidney, they leapt at the chance to compete. MIMETAS is one of several companies that have arisen in recent years to create “organs-on-chips,” cell cultures grown in such a way as to recreate the three-dimensional physiology of living tissues. (See also “Cambridge-Based Emulate Aims to Jumpstart Organ-on-a-Chip Market.”) The best organs-on-chips may have two or more cell types from an organ connected together, fed by vascular tissue through which fluids flow naturally through the system. They can also use cells from specific subpopulations of patients, or different disease states, to refine testing.
“What you can do with organ-on-a-chip models is way beyond what you can do with animals,” says Joore. “You can use human cells from a whole range of different backgrounds. Geographical backgrounds, racial backgrounds. You can add into a model the complexity of the human population.”
MIMETAS joined the CRACK IT NephroTube Challenge late last year, partnering with a group from the Radboud University Nijmegen Medical Center who had developed the ciPTEC kidney cell line. CiPTEC cells are derived from the human kidney endothelium, and have a mutation allowing them to quickly expand to a controlled size, at which point they begin to express the ion transporters used in the human kidney for absorption and excretion. “It’s a very nice, very elegant model,” says Joore.
Less than a year later, he adds, the kidney-on-a-chip developed by MIMETAS and the Nijmegen team “is the most physiologically advanced [model] we have. It contains an endothelial tubule, some connective tissue, as well as a proximal epithelial tubule. Taken together, this is a micro version of a proximal tubule in your kidney that can mimic most of the transport processes.” The judges of the NephroTube Challenge, including representatives from GlaxoSmithKline, Pfizer and Roche, awarded the MIMETAS team the full £1 million prize this September, funding that will go toward further developing, characterizing and validating their kidney model.
A Fluid Enterprise
MIMETAS began not with a vision for animal-free drug testing, but with a new method for handling liquids in small chambers. Between 2003 and 2009, Paul Vulto, working at Silicon Biosystems in Bologna, Italy and the IMTEK institute in Freiburg, Germany, invented a structure called a phaseguide, a patterned barrier that can be placed between two chambers in a microfluidics system. Phaseguides have the important property that a liquid will not cross them until it has completely filled a chamber, at which point they become permeable and let liquid into the next chamber. This makes them useful as both valves and “train rails” in microfluidic chips.
Four years ago, Vulto had a dinner meeting with Jos Joore to show him the technology, and the two immediately decided there was a commercial market for phaseguides in the emerging lab-on-a-chip field. “We lined up five or ten different applications,” Joore remembers, “and we were so enthusiastic that we decided to start MIMETAS on the spot. We used the next six months to find the application where we expected the biggest added value, and the biggest market need, and that was in 3D cell culture.” Joore became chief business officer of the new company, and Vulto became chief technology officer.
Using phaseguides for cell culturing allowed the newly-formed MIMETAS to layer fluids containing human cells beside one another in separate lanes, letting tissues grow independently, but also freeing them to interact across channels. A lane could also act as a sort of artificial bloodstream, creating a flow of growth medium and nutrients through the tissue, which could also be used to introduce drugs for preclinical studies.
Co-founders of MIMETAS Paul Vulto (L) and Jos Joore. Image credit: MIMETAS
Today, MIMETAS sells both general-purpose and custom organs-on-chips built on top of its OrganoPlate platform — a gel plate engineered with dozens of microfluidic culture chambers, each of which contains two to five parallel lanes into which different cultures and media can be introduced. The aim is to make organs-on-chips into mass produced assets for use in high-throughput drug screening. “Once you have robust culturing conditions, it is fairly straightforward to fill up, by robot or even manually, a hundred plates or more, which would provide you with 10,000 data points,” says Joore.
Getting to that point, however, is rarely a smooth road. While every new organ model is built inside similar “chips,” different tissue types have their unique demands and complications. First, MIMETAS must sort out which cell types are essential to an organ model, and find the best cell sources to represent them. Joore gives the example of creating a lung-on-a-chip for studying asthma. Lungs have different epithelial cells and fibroblasts that form the structure of the tissue, blood vessels to supply it with oxygen and nutrients, and immune cells moving through the system. MIMETAS’ job is to discover which of these are essential when working with specific diseases; for asthma, an epithelial boundary, a blood vessel and immune cells are a good start.
Then there are the physical constraints of the organ. Some tissue cultures, like those from the intestine, want to grow fairly large, up to a millimeter in diameter — but changing the volume of the microfluidic channels in an organ-on-a-chip also affects the flow rate, and the ability of the tissue to take in oxygen. Interconnected cell types also need to grow to sizes proportionate to one another, if the boundary between them is to mimic the absorption properties of a real organ. “The basic technology, the OrganoPlate, is fairly easy to work with,” Joore says. “The challenges we face are mostly biological challenges.”
Those challenges only multiply as the models get more specific. MIMETAS distinguishes between its “precompetitive” products — generic organs-on-chips it sells to all customers for tox screens and pharmacokinetics studies — and the “competitive” products it develops one-on-one with particular partners.
Competitive models are usually disease-specific, and they can involve some elaborate engineering. “Our disease models are a mixed batch,” says Joore. “Because we need relatively small amounts of tissue, we are very well equipped to work with primary cell types,” derived directly from patients. These tend to be more loyal to the true biology of a disease than cell lines, but are often harder to use because they are tricky to acquire and have limited lifespans.
In other cases, MIMETAS uses mutant cell lines, which are genetically programmed to express a certain disorder. “Cancer is a different ballgame altogether,” adds Joore, “because cancers are often immortalized as they are,” making cell lines more reliable. In every case, the company must be careful that tissues grown in OrganoPlates actually express the correct disease phenotypes once they grow into 3D structures, which can be a painstaking process of trial and error.
Growth Markets
In collaboration with the Erasmus Medical Center, one of MIMETAS’ leading projects is growing a series of neurons-on-chips. Erasmus supplies patient-derived cells for the program, along with a genotype of the patient the cells come from, giving MIMETAS a wealth of information on the biological sources it’s working with.
Growing neurons in an OrganoPlate, and letting the cells differentiate, takes around two months. “You get neurons that are electrically active,” says Joore. “They’re excitable and even have some cell activating activity, so they’re fully physiologically relevant and functional.” The collaborators have worked hard to characterize the activity of their neurons-on-chips, showing, for instance, that the cells express synapsin proteins in ways that are typical of neurons with working synaptic contacts.
“Now that we have neurons, we want them to express a disease phenotype that presents our customers with a screenable model,” Joore adds. That’s the real core of MIMETAS’ business, and while the organ-on-a-chip market is still very young, it’s beginning to show commercial results. MIMETAS has customers in both the large pharmaceutical companies, including Roche and AstraZeneca, and smaller biotechs and academic groups. Projects like the NephroTube Challenge also give the company an early glimpse into what qualities will make organs-on-chips attractive to the major drug companies looking for new ways to test their candidate compounds.
OrganoPlates, MIMETAS' organ-on-a-chip platforms, showing culture lanes with phaseguides. Image credit: MIMETAS
MIMETAS also benefits from local support in the Netherlands, where its headquarters city of Leiden is a regional biotechnology hub. One of MIMETAS’ early partners is Galapagos, a Leiden-based company that pursues internally discovered drug targets and has been aggressive in using novel cellular models of disease. “Galapagos is a highly innovative screening company, and was daring enough to take the leap to use unique models in order to find unique molecules,” says Joore. “We make very specific, defined disease models for them, allowing them to screen smart libraries.”
Dutch investors also led a $5.2 million funding round for MIMETAS earlier this month, to scale up manufacturing and inject new resources into the company’s development pipeline. The two venture capital funds that led the round are PPM Oost, one of the Netherlands’ largest investment firms, and Zeeuws Investerings Fonds, a fund run by CEOs of multinational companies who also advise on international business expansion.
The future of organs-on-chips is still uncertain, and huge amounts of data will have to be gathered on their reliability before they can gain the same broad acceptance as animal models or cell lines. But like its competitors, MIMETAS suspects the time is ripe for serious changes in preclinical testing, as expensive failures in clinical trials continue to mount up. An early entrance into the organ-on-a-chip market may prove to be a crucial edge later on, as MIMETAS begins to measure its models’ performance in drug screens. The company’s kidney and liver models are both going through forward validation, where libraries of failed drugs are screened again to see if the organs-on-chips could have caught issues early on.
The precompetitive OrganoPlate models are an important asset in creating a market for organs-on-chips. By getting them into multiple companies’ hands, MIMETAS hopes to generate a baseline of toxicity and pharmacokinetics data, which could encourage new customers to give the models a chance. “We think it’s good for everybody if they’re used as broadly as possible,” says Joore. “That provides better validation, so the product gets better the more it’s being used.”
Joore will be visiting Boston next month to speak about MIMETAS’ OrganoPlate technology, at the Screening and Functional Analysis of 3D Models conference on November 18 and 19.
