In this Founder Stories presentation, Armon Sharei, co‑founder and CEO of SQZ Biotechnologies, outlines the evolution of therapeutic modalities from small‑molecule drugs to biologics and now to cell‑based therapies, emphasizing the central delivery bottleneck that limits cell engineering. He introduces SQZ’s microfluidic “cell‑squeezing” chip, which mechanically deforms cells as they pass through a constriction, temporarily disrupting the plasma membrane and allowing a wide range of cargo—DNA, RNA, proteins, nanoparticles—to diffuse into the cytosol. The device operates at ~1 million cells per second with high post‑process viability, outperforming traditional methods such as nanoparticle uptake, electroporation, and viral transduction. Sharei presents pre‑clinical data showing that antigen‑presenting cells processed through the chip can present tumor antigens, elicit robust CD8⁺ T‑cell responses, and mediate tumor regression in mice. He discusses broader applications, including reprogramming skin cells to induced pluripotent stem cells, and a vision of autologous immune‑cell engineering for cancer, infectious disease, and regenerative medicine. The talk also covers the company’s timeline (idea in 2008, discovery 2009, team growth to ~15 by 2014, now a commercial biotech), recognitions (Scientific American Top 10 World‑Changing Ideas 2014, Forbes 30 Under 30, Fierce 15), and leadership (board of MIT professors, executive chair Amy Schulman). The presentation ends with a brief Q&A about the serendipitous origin of the technology, development timeline, and future therapeutic breakthroughs.
Chapters
Our next and final presenter for this session is co-founder and CEO of SQZ Biotechnologies, Armen Sharey. Dr. Armen Sharey received his BS from Stanford University and his PhD in chemical engineering from MIT under the guidance of Clarence Klaus Jensen and Robert Langer. He was a national finalist for the NIH's Director's Early Independence Award in 2013 and a Reagan Institute fellow at Harvard Medical School. Co-founder and CEO of this, of this biotechnological company, he was named one of Forbes' 30 Under 30 in healthcare. The work of his team in developing the microfluidic methods for robust intra-cellular Cellular Delivery of Materials for Research and Clinical Applications, was named one of Scientific America's top 10 world-changing ideas of 2014. He has co-authored more than 13 peer-reviewed publications. His inventions have received 6 patents. Watch out for his upcoming appearance in The Big Bang Theory. No, I'm just kidding. Ladies and gentlemen, Armen.
Now I feel like I should have prepared some Big Bang Theory jokes, but I don't have them. How do I pull this up? Aha, magic. All right. So I'm going to talk today about squeezing cells and why you should care about that. This is done on a chip that's about this big, so about the size of Spock's communicator thing. Actually a little smaller. But why would you want to be putting these scared little cells through a constriction like that to put stuff inside them? So if you think about how drugs work, most of the drugs you deal with are what we call small molecule drugs. So these are synthetic chemicals, sometimes of natural origin, that tend to mimic the behavior of certain biological molecules in your body. And by interacting with various things inside your body, they can have a therapeutic effect. And this was kind of the main type of drug people were using for a long time until people like Genentech and Amgen came along and developed what we call biologics, which are molecules that very much mimic or are a slight modification of what your body actually has inside it. So peptides, proteins, etc., materials that your body would naturally produce. And these drugs started to give much more efficacy in certain applications because they were a lot more similar to what your body had, and therefore they could be a lot more specific or address a lot of diseases that these small molecules weren't working for. And now there's a trend towards using entire cells for therapy. So as a sense of scale, this cell is thousands or hundreds of thousands of times larger than any of the proteins that make it up. And these are really biological machines that do everything that our bodies do. And the idea is that if you can start to kind of take control of these biological machines, you can start to have much more effective therapies for things that are challenging like cancer or HIV or Ebola. And when you think back to the history of what happened between going from small molecules to biologics, A lot of this revolved around the development of recombinant DNA technology, which allowed people to actually produce these proteins to then go and inject into patients. And then a big question in the field has been, okay, now how do you make the next leap from biologics to cell therapies? Because it's been very difficult to engineer cells effectively to fight your disease for you. Now the way we kind of look at this is more of a delivery problem. So in the end, in order to engineer a cell, you need to be able to get stuff into inside it to modify its function. And there's a number of ways to do this currently. So these disparaging cartoons help illustrate how we feel about what can currently be done. So there's methods wherein people produce nanoparticles that try to find their way into your unsuspecting cell. Cells typically tend to not like this, and some of these particles don't work very well. Another way is to zap them with an electrical field to try to get your stuff inside. They also don't like getting electrocuted. And you can also try to engineer viruses that are typically, you know, your flu virus or modifications of an HIV virus to get stuff in. These all have their own issues and have largely limited what's been possible. And the way we think about the challenge of getting materials into a cell is kind of along these two axes. So one is the material you're trying to get in. And I'll spare you the details about these molecules, but largely there is, you know, nucleic acids like DNA and RNA, and then there's everything else, so proteins, other synthetic molecules. And then you can think about the type of cell you're trying to put it into. So you have cell lines, which is what people grow in the lab and do experiments on. They're not really representative of the cells that are inside us, but they're kind of the closest thing you have in a lab. And then there's patient-derived cells, like the cells that make up you and I. And when you look at what's possible with these existing systems, the only thing that you can really do is deliver this subset of materials into these model cell lines. If you try to deliver those materials into primary cells that you derive from us, it doesn't work that well. And then there's whole classes of materials that you can't put into these cells. And this kind of map shows how limited we are and what we can get into these cells to actually start to engineer them. So what we developed is these little chips that I showed you. And the way these chips work is that normally this is now zooming into a cell. A cell has a membrane that prevents stuff from going inside it. And typically that membrane is very, very good at preventing anything from the outside getting in. And what we've done is in these chips that I showed you, we have cells that go through channels at high speeds. And as they're going through, they go through this constriction point which rapidly deforms them. And that deformation results in the temporary disruption of the membrane so that anything that's outside can diffuse in. And then these cells manage to seal it back up after about 15 seconds to 2 minutes, depending on how you're doing it. And this little chip can run about 1 million cells per second, so it's actually not an insignificant amount. Of stuff that you can pump through such a small chip. Now thinking back to this rubric of, you know, these areas that have been very difficult to deliver material into, we've done work in all four areas of this matrix, and what we've seen is that we can actually get this system to work in areas that previously were challenging. So in the green zone where current methods work, this system works as well. When we start to look at the yellow zone, we start to see significant improvements upon what exists. So for example, the cells that go through these devices can be a lot more viable than the alternative. And if you have happy cells, they're more likely to do what you're asking them to. And there's whole classes of materials that we can deliver now that you couldn't have gotten in before. So these are certain nanoparticles that can be engineered to serve functions inside the cells.. And in other areas we showed that we can actually reprogram the function of your cells 10 to 100 times more effectively than any other method. In this particular case, we were taking skin cells and turning them into stem cells, which can be particularly useful for certain applications like regenerating a severed spinal cord or something. Now, the ultimate vision is really that there's a lot of molecules that people know of or Classes of molecules that people know of that if you could get them into the cell, they could start to affect significant changes in those cells and make them start to do what it is you want. And so if you could actually facilitate delivery of all these different materials, can you now start to engineer these cells to do whatever it is you want? So I want a cell that's going to home to the brain and start secreting an anti-Parkinson's factor, for example. These are all things that you can start to envison if you can in fact engineer these cells. And ultimately down the line, kind of the workflow that we would want to implement is a situation where you have your cell engineering materials and then you take blood from your patients, which has their happy little immune cells in it, run them through one of the devices to deliver those cell engineering materials. This will in turn train them against the disease, and then you can put them back into the patient. And this has generated a lot of excitement. So for example, the NIH was quite excited when we were first able to demonstrate some of the data that we did. And one of their directors had a nice quote for us, which was, this deceptively simple new way to control cell behavior offers exciting promise for studies of basic biology, as well as enabling cell-based therapies previously only envisioned. To bore you with the science a little bit on an example of where we've seen some of these differences. We'll do a quick run-through of some Immunology 101, which is if you think about cells, even tumor cells, they all have an interesting mechanism by which they take proteins that are inside them, so these big squiggles, and they break them down and present small pieces of them on their surface. This is the mechanism by which your cells basically show the rest of the immune system, here's what's inside me, if you detect anything that's bad, please kill me. And the cells that come along and do the killing are your CD8 T cells, which are very powerful assassins. So these are the T cells that save you from most infections, and they've been shown in people that can control HIV or people who have very good tumor responses. Cells are very active and very much responsible for helping to protect you. And these cells are extremely good at their job, so they will only attack cells that present exactly the peptide or fragment that they want to see. And if they see it, they'll secrete materials that'll kill the cells. So now the question is, how do you get these guys, these very capable assassins? The way you get them is that you actually have The exact same fragment has to get presented to them by a different type of cell called an antigen-presenting cell, which I won't go into detail as to what's exactly different about these guys, but they're very effective at inducing responses with these T cells. So when these T cells see their fragment on their, let's call it, command and control cells, they will then proliferate, expand throughout the body, and go hunt down any cell that matches that profile and they'll kill it. Now the idea has been, can you, by manipulating these antigen-presenting cells, start to train our own immune cells to go and fight your disease for you, for example, in the case of a tumor? And so what we did is that we took these antigen-presenting cells and we asked the question, okay, if we could take these target proteins that we know are, for example, tumor-associated, and we used our device to disrupt the cell's membrane so they can go in. Can you now get these cells to process and present the fragments that you want? And can these fragments then be shown to CD8 T cells? And will they in turn respond and proliferate? And really, is all it takes just to disrupt the membrane? So this is a complex biological process that happens in various parts of your body. But if we could just do step 1 and put these cells back in, will your body do the rest? And at least, or sorry, and once you have these CD8 T cell responses, can they then go and destroy your tumor? And what we've been able to show so far is that at least in mice, this does in fact work. So if you do try to engineer your antigen presenting cells and put them back into a mouse, you can start to see massive T cell responses in these mice. As compared to if you did nothing, or what kind of the next best thing is in the field. And these responses can really potentially start to enable cures wherein you're just using your body's own immune system to fight the disease for you. Now, from a scientific perspective, I guess you stole all my highlights, but there's definitely excitement in the field about what we could potentially accomplish. So Scientific American named us as one of the top 10 World Changing Ideas for 2014. Recently, we were very proud to be recognized as one of the top 15 private biotech companies as part of the Fierce 15. And we've had a number of publications and other recognitions with our collaborators. So a lot of pressure for us to actually pull this off. And the team that's going to help us do that is our board members are mostly MIT-associated professors, and as well as our executive chair, Amy Schulman, who was a Pfizer executive actually until last year when she decided Boston was better than New York for some reason. She's not a Red Sox fan yet. And we have our leadership team at the company, so it's been very exciting to build this up and try to move forward this broader vision of actually enabling cell therapies. And hopefully by addressing that fundamental problem of getting material into a cell, we can now help engineer a whole generation of therapies where you're using your own cells to fight the diseases for you.
Very good. Great. All right, the first question that pops into my mind— beyond, did you get all A's in biology in high school?
Biology was actually not one of my favorite subjects. Really? I like math and, like, physics.
Okay, because I felt like I was in biology class again. So what's been the most important innovation that's enabled this technology?
I think it was really the somewhat serendipitous finding with the chip. So, you know, squeezing a cell doesn't really sound like the best idea up front, and it wasn't necessarily what we set out to do. What we were hoping to do was actually develop a gun that could fire fire things into a cell. And that makes logical sense, but practically wasn't really working. And through the course of trying to develop a system that was better at shooting material into a cell, we kind of came to this finding that if you just squeeze a cell hard enough, you can disrupt its membrane and anything will go inside. And that was really kind of the fundamental breakthrough that fueled this, because it was also so simple that it really allowed us to apply it across a broad range of areas.
When you embark on something like this, how— give us an idea of how long this usually takes, this process. How many folks are involved? Is it in a lab? Are you going to code? I mean, what's it look like?
I think it looks different for every group, but to give you a sense, we first— so this project started in around 2008. We came to the realization with the squeezing around 2000, mid-2009, and At that point, it was just 4 or 5 people, so me and a troop of undergrads. And eventually that started to build momentum. It was probably like 15 people or so working on it by 2013, 2014. And now there's the company as well as a lot of collaborators around the world. So it definitely takes a while.
And what are some of the major medical breakthroughs that could stem out of this in the near future?
I think one of our biggest, most optimistic targets is the idea of taking your immune cells and engineering them against cancer. So taking your immune cells, training them what the tumor looks like, and then using those CD8 T cells as assassins to go out there and destroy the tumor for you. And there's already evidence in the clinic with other means that if you could accomplish this, you actually might be able to get even cures. In some cases.
Great. We look forward to that. All right. Thank you. Thank you. Good luck.
