Engineers Are Reinventing Drug Testing
A Techwise Conversation with Nina Tandon, EE and tissue engineer
Steven Cherry: Hi, this is Steven Cherry for IEEE Spectrum’s “Techwise Conversations.”
Testing newly developed drugs is utterly necessary, but it’s not without its problems. For one thing, it’s expensive. For another, it’s not very accurate—it usually starts with mice, and they’re not like people in the ways we need them to be. It continues with humans, and often even human subjects are not like the people needing the drug, in the ways we need them to be. And we’re putting those people at risk. And did I mention how expensive it is?
Here’s a kind of fantasy scenario: We grow stem cells and make them mimic whatever we want. Then we hook them up to microprocessors so we can study them better. Let’s take the fantasy one step further. Organs and drugs interact in our bodies, so let’s make a bunch of different kinds of cells, and now hook them all together so we can mimic an entire organism. Oh, and if you could make them out of something that isn’t a stem cell, with all the attendant controversy, that would be even better.
My guest today is one of the leaders of a wave of researchers doing just these very things. Nina Tandon is an electrical engineer working with a new type of cell, called an induced pluripotent stem cell, for drug trials and other biomedical research. She started out in telecommunications and would be able to explain, if we wanted her to, how the same equations govern cable transmissions across the Atlantic Ocean and nerve transmissions across our bodies. She’s a research scientist at Columbia University, was a 2011 TED Fellow—and had an awesome TED talk in December, on which I’m shamelessly basing my questions—and she’s an avid member of the IEEE’s Engineering in Medicine & Biology Society, or EMBS. She joins us by phone.
Nina, welcome to the podcast.
Nina Tandon: Good morning. Nice to be here.
Steven Cherry: Nina, I guess we should start with induced pluripotent stem cells. What are they?
Nina Tandon: They’re kind of like stem cells that have been tricked into being stem cells. You take cells, say, from skin and transect them with just a few genes, about four genes, and you can trick the cells into going into an embryonic state. And then the cells, I mean, they’re not exactly embryonic cells, but they behave in many ways similar to embryonic cells, in that you can get them to differentiate into all kinds of different cells that you would find in the body. And so because of that, they’re really great. They’re like embryonic cells without the controversy associated with human embryonic stem cells, but then also you can generate embryonic-like stem cells from any living person as an adult.
Steven Cherry: So now, how are these hooked up to computer chips?
Nina Tandon: Well, this is a new area of research for us. I started my Ph.D. growing cardiac tissues, and cardiac tissues, they grow really well if you subject them to electrical signaling. I started using as part of this research what are called “microelectrode arrays.” And these are printed electrodes that can measure, kind of like mini-EKGs, from the cells that are growing on top of them. And so we can, you know, using lots of amplifiers, we can get these low-power signals out of the cells and then begin to perform analysis on the signals.
Steven Cherry: So you simulate a drug trial by growing thousands of them, as if you had thousands of patients.
Nina Tandon: Yeah, so this is something that I’d like to do going forward. We’ve been able to measure signals out of individual pieces of tissue, but scaling up into, you know, the thousands is something that we’re looking to partners in the electronics space in order to do. And by the way, electronics companies, I’m not going to name names, but most have been reaching out to us to do this. So I think there’s a lot of interest both from the research side as well as from the technology side in scaling up this kind of technology.
Steven Cherry: So is it one microprocessor per cell, or is it multiple cells would be hooked up to a single microprocessor?
Nina Tandon: So we’re growing mini tissues, so there would be thousands of cells in each of the microtissues. And each of those kind of little blobs of heart need to have their own amplification and their own analysis associated with them. So you can architect the system however you like, but those are the dedicated resources you’d need per microtissue.
Steven Cherry: Very good. So there’s a lab at Harvard that’s used this strategy to start studying Lou Gehrig’s disease?
Nina Tandon: Yeah. Kevin Eggan’s lab up at Harvard has, and actually he’s also [chief scientific officer] at the New York Stem Cell Foundation. His lab has done something really interesting: So they took skin cells from people who had the gene for Lou Gehrig’s disease and they engineered these induced, these pluripotent stem cells out of them, and then differentiated those cells into neurons. So it’s kind of like taking skin, turning it into embryonic cells, and then turning those embryonic cells into neurons. And what’s interesting is that those cells actually show some symptoms of the disease. And so it’s a great way to be able to study ALS, Lou Gehrig’s disease, and in vitro in the dish, in the lab.
Steven Cherry: So the ultimate goal is to build this up into an entire ecosystem that simulates the body, right? So that you get the reactions between, I guess, what happens in the kidneys and the pancreas and the thalamus gland, for example, or bone marrow and autoimmune disorder.
Nina Tandon: Yep, sure.
Steven Cherry: How far away from that are we?
Nina Tandon: There are lots of different labs working on this, actually, and there’s a nice competitive NIH grant out there for groups to develop these kinds of chips. And Don Ingram’s lab is one that’s really pioneered the stuff of mixing tissues on the same chip [including a lung-on-a-chip, gut-on-a-chip, and human-on-a-chip].
So there are several examples that are really cool so far. There’s, I think there’s a gut chip that actually has microbes on it. And lung on a chip. Combining different tissues together is kind of just starting now. Our lab, our supervisor, and two collaborators, one at Harvard and one at UPenn, we’re working together now…so our collaborators, Sangeeta Bhatia and Chris Chen. Sangeeta at MIT, her specialty is liver. She’s been growing liver for 10 years. And Chris Chen grows endothelial cells. We’re really good at growing cardiac cells, so we’re all working together to try and get these tissues on the same chip.
We just got this grant, literally a month ago, and so the next two years are going to be really crazy as we try and combine our technologies together, but it’s going to be a lot of fun. A lot of drugs that fail do so due to cardiotoxicity. And a lot of times there’s what’s called idiosyncratic cardiotoxicity that has to do with by-products from the liver. And so growing these tissues on the same platform is going to begin to enlighten us in all kinds of ways we hopefully don’t know about yet.
Steven Cherry: Now, another direction you would like to take this in is personalized, individualized drug testing.
Nina Tandon: Yes.
Steven Cherry: I guess the idea is, if you’re going to start out with a breast cancer cell, why not make it my breast cancer cell?
Nina Tandon: Yeah, that’s the idea. We’ve been seeing hints at personalized medicine in cancer for a while. There are a lot of genetic tests. People take the genome of their cancers, not almost routinely, but quite often now, because it’s known that certain mutations in a cancer will make them more or less likely to respond to a drug. Going forward, I think it would be even more useful to be able to culture, to combine that knowledge with cultured cancer in the dish, to be able to test drugs and test cocktails of drugs on them.
Steven Cherry: Yeah, so how far down the road is all this? I mean, when do you think this technology would actually start to affect drug testing?
Nina Tandon: Actual drug testing is, I think it’s already happening. I think oncology is one of the pioneers in this space, just because it’s been, you know, biologics and biologically based drugs have quite a long history with oncology. I think we’re, let’s see, I remember reading that there are actually existing mouse models, so people are taking their cancer and injecting them into mice, and treating those mice with cancer drugs. Right now, that’s happening, to be able to predict how those cancers will react to certain drugs.
Because you don’t have a lot of wiggle room, if someone has a really dangerous cancer, to be able to experiment on them, right? So we’re already seeing some of this stuff happen in cancer. I think it’s going to accelerate, but I think cancer’s really kind of ahead of the game with pioneering some of these methods.
Steven Cherry: And when do you think we would start seeing drug trials that involve the induced pluripotent stem cell and computer-chip-type testing?
Nina Tandon: I think that could be soon. I think that could be in the next five years, honestly. I think you’ve already seen…so they’re based on methods that already exist. So electrophysiological methods are common for measuring responses in the heart, right? And responses in cultured heart tissue. I think scaling them up and using big data methods, basically, using high-throughput screening, I think we’ll start to see soon, because it’s more of a technology barrier as opposed to a psychological barrier, you know, of not wanting to use that kind of evidence.
I think those imply different strategies for working around the issue. Now that said, I think to be able to increase the psychological trust that people place in these kinds of results, I think that our field is going to have to get better at demonstrating uniformity of tissues. So that scalability on the tissue side, as opposed to the electrophysiology side, is going to be an important piece. And that’s not easy. We culture embryonic stem cells, and we still get variability in how many of those cells start beating, for example. We have to get better at that before we can expect people to trust the results of engineered tissues in place of animal models.
Steven Cherry: And I guess my last question for you is, do you ever have a feeling that there’s a kind of Frankenstein aspect to this? You’re kind of growing these sort of body parts and bodily ecosystems in a lab, and hooking them up to computer chips and then starting to manipulate them in various ways.
Nina Tandon: So, I have been called Dr. Frankenstein, like, at cocktail parties, sometimes. Not cocktail parties, but, you know, jokingly so, because it’s rare to find an electrical engineer working in electrical stimulation of tissues in the lab. There’s not a huge, you know, population of us. So I think that people draw that analogy. That said, do I think of it that way? I think it doesn’t feel Frankenstein to me, but it does fill me with appreciation for these cells. It makes me, I can’t help but think those same cells, those same kinds of cells, are inside our bodies every day and that they don’t need that much to do what they do best.
So I don’t necessarily think of myself as shocking the cells to life. I think of myself as learning to speak with cells that are already alive. It’s a subtle distinction, but to me I don’t think of these cells as not being able to live without my intervention. And I think that’s what Frankenstein is about. I think of these cells as living, beautiful things, and that I’m building technology to help speak their language. So, no, it doesn’t feel like Frankenstein to me. Does that make sense?
Steven Cherry: I think it makes perfect sense, and I also think that to study something in the depth that you are is also to honor it. So on behalf of all of us who need drugs and need drug testing, and need for it to be as accurate as possible, I think this is really amazing research, and I thank you for doing it.
Nina Tandon: Oh, thank you so much. Thank you for the opportunity to speak about it. It’s really fun. It’s been a pleasure.
Steven Cherry: We’ve been speaking with Nina Tandon about how she and other scientists are hooking up human cells to computer chips, for drug testing and other medical research.
For IEEE Spectrum’s “Techwise Conversations,” I’m Steven Cherry.
NOTE: Transcripts are created for the convenience of our readers and listeners and may not perfectly match their associated interviews and narratives. The authoritative record of IEEE Spectrum’s audio programming is the audio version.