Flagship Pioneering: Avak Kahvejian, General Partner
“If you are trying to build something totally new, and be a pioneer, then science becomes the most important thing.”
Staring at the pipettes, agarose gels, bacterial cultures and worn molecular cloning protocols strewn across his lab bench, the newly minted Dr. Avak Kahvejian realized three things about himself.
First and foremost, he liked doing difficult things. It is what led Kahvejian to major in molecular biology rather than computer science at McGill—mapping the features of biological life seemed harder than debugging algorithms (Q#1). If given the choice, he would take the less well-trodden path, the riskier bet. His desire to work on the “bleeding edge” of knowledge led him to join the Sonenberg Lab as a graduate student. There he felt like he was “really doing science” by exposing the intricacies of how eukaryotic cells turn RNA into protein—a crucial step in the central dogma.
However, his excitement in the lab was dampened by his second realization: he yearned to be part of the scientific discoveries splashed across the news in the early 2000s. Yeast cultures and DNA plasmids in hand, he felt isolated from the breakthroughs: “Francis Collins and Craig Venter were battling it out in the public sphere, with Bill Clinton in the middle of it. The genome revolution was opening up new avenues and opportunities for biology and biotechnology” (Q#2). Feeling that academia was not for him, Kahvejian set out to find his true calling—he talked to scientists in biotech, instrument and tools providers, scientific liaisons for big pharma and business development associates.
After briefly considering starting his own company (Noubar Afeyan told him it “made no sense” Q#3), Kahvejian’s exploration led him to legendary entrepreneur Stan Lapidus. Given 48 hours to accept a job offer, he took the plunge, moved to Boston and joined Lapidus at Helicos. The company was building the world’s first single molecule DNA-sequencer. It was a bold and complex project—one that was under constant threat from the economic downturn of 2008 and competition from market leader, Illumina.
Jumping from business development to product management to R&D and back to business development, crystalized Kahvejian’s third realization, one that he had perhaps sensed from the start: he was a born entrepreneur. He relished taking on roles that he “didn’t know existed” the day prior, tackling grand problems, and building teams from the ground up. Though Helicos succeeded in making an extraordinary sequencing technology, it failed to gain market penetration. As Helicos was contemplating a pivot from selling research tools to developing and performing diagnostic tests, Kahvejian was exposed to Flagship Pioneering’s innovation process: “At this point, I got a lot more exposure to how Flagship works. Noubar was on the board, and together we started to ideate about how we could apply the [Helicos] technology to other verticals and markets… Flagship has much more of an entrepreneurial or innovator’s mentality versus one of an investor.” (Q#3)
In 2011, Kahvejian jumped to Flagship Pioneering as a partner. Over the past 11 years at Flagship, Kahvejian has led an innovation team that has created multiple public and private companies, including Seres Therapeutics (NASDAQ: MCRB), Rubius Therapeutics (NASDAQ: RUBY), Codiak BioSciences, Cygnal Therapeutics (now Sonata Therapeutics), Ring Therapeutics, Cellarity, Laronde and Generate Biomedicines.
Microbiome therapeutics, cell engineering, design of novel gene therapy and RNA platforms, effective use of computation for biological discovery—these are just a few of the diverse and challenging concepts that Kahvejian and his team wrestle with daily: “At Flagship you are perpetually building a yet unmade future…I love it.”
I think it is fair to say that the once restless grad student, always probing for the most difficult path, has finally found his home building innovative bioplatform companies.
Below is an interview with Dr. Avak Kahvejian from September 2022:
1. What first got you interested in science and medicine?
I've always been interested in science. When I was a kid, I loved science fiction, space exploration, computer programming and biotech, insofar as I understood it [at that age]. I also have a bad personality trait, which is if I think I understand something, or that it is “safe” or “easy,” I'll always go for the harder thing, the less well-trodden path. When I was picking a major in college, I was originally thinking about computer science, but I realized I wanted to pursue something that's even more at the edge of human ingenuity and knowledge, which led me to biotechnology.
[At McGill] I learned biochemistry, molecular biology, and cell biology, and I was really enthused to study these disciplines. But I later became a bit disillusioned in undergrad, because it felt as if we were just memorizing a bunch of stuff. It was cool stuff, but I realized I wanted to be more lab oriented and experimental. [Doing research] made me feel like I was really “doing science” and was on the bleeding edge—not just memorizing but creating new knowledge.
So, I joined the Sonenberg Lab at McGill. He [Nahum Sonenberg] was well-renowned and was working on understanding how mRNA translation really works. I thought that it sounded like a great place to be. When I joined his lab [for a PhD program] I didn’t have an eye toward becoming an academic or professor. I was just pursuing my curiosity and continuing to go into the more challenging or unknown areas.
[On his PhD work in the Sonenberg lab at McGill]
We were doing super basic research on the eukaryotic translation factors that bring the ribosome to mRNA, or in some cases, inhibit ribosome binding to the mRNA. As you know, ribosomes are big molecular machines that can read mRNA and “translate” it into polypeptides and proteins.
Each member of the lab was working on one translation factor, and trying to characterize its function, regulation, and involvement in higher-order processes. For some translation factors, we knew enough about their binding and regulation such that people were looking at their involvement in disease or physiology. For other translation factors, we had just recently cloned and knew nothing. In this case, we had to do the basic work of finding out where it binds [to ribosomes and RNA or to other translation factors], how these factors are post-translationally modified, and what the protein does to regulate translation. In the beginning this was exciting, but eventually I couldn't believe that I was working on such a narrow piece of the entire puzzle.
2. At what point did you consider a "non-traditional" career path outside of academia? Was there a mentor that inspired or supported you to do so?
After some time in grad school, I got a bit disillusioned again. I thought the research was cool, and I loved manipulating biology and trying to understand it, but I didn’t feel like I was going to have a big impact if I just identified the phosphorylation pathway for ‘PABP interacting protein 2’, which is literally the name of the protein I was studying.
The genome revolution was happening at the time [early 2000s]. I was reading the news and comparing what was going on to what I was doing at the bench. I realized that I was in the field of science, but it didn’t feel like I was in the field. Lab work didn’t feel as exciting as what was going on in the news: Francis Collins and Craig Venter were battling it out in the public sphere, with Bill Clinton in the middle of it. The genome revolution was opening up new avenues and opportunities for biology and biotechnology. There were all of these cool machines and computational tools used to sequence and assemble genomes emerging. Meanwhile, I was sitting there with gloves, a gel and yeast or bacterial cultures. I wanted to be part of that world and have a broader impact on humanity.
That feeling is what got me asking: How do I really do this? What should I do next? Do I try to become a professor with my own lab? Or do I join the biotechnology industry? I was still quite naive and ignorant, and so I started to explore the alternatives [to academia]. Initially, I painted industry all with one brush and academia with another brush, with no in-between. Industry seemed like a monolithic thing at first, but it turned out that there were many scientific, commercial, and financial roles, as well as players in the industry. I had to educate myself on the landscape of big pharma, tools providers, and small startups.
As a PhD student at the time, we had no exposure to these different jobs. We didn't really interact with startups at McGill. At the time [in Montreal] there were only one or two biotech startups. I knew about them, and some of the professors were involved, but that [industry involvement] was not very visible. So, I had to kind of build that map for myself by literally walking around, meeting people, and making connections: “Oh, this person's mom works at Pfizer. Let me go ask her about her job and what Pfizer does. This person got a job working at Boehringer Ingelheim. What does she do? Let me go talk to her.” I realized that over time the startup world was most exciting and compelling for me.
It was the first time I learned to take advice with a grain of salt. I needed to find the kernel of truth in what people were saying, as well as ignore to some degree what didn't fully sit with me. This includes your parents, right? My dad told me to go be a professor because of the high likelihood for stability. He advised that I could continue to do what I liked and succeed, and that by going into biotech, I would be throwing it all away and starting from scratch. For me, I didn't feel that this type of thinking made any sense.
3. What was your first "business" experience out of grad school? What did you learn?
When I was finishing my PhD, my professor got approached to launch a startup. I thought this was my ticket to working for a startup at the cutting-edge of science. I could be one of the first people involved to get it off the ground. But it was a terrible idea, and a terrible situation in terms of who was approaching us.
Even so, it was tantalizing and exciting. To me, this really was a signal that I was much more excited by startups [than by other avenues]. I actually approached Noubar [Afeyan] with the startup idea to ask his advice. My family knew him through our Armenian connection. When I brought him the idea he said: “What are you talking about? This makes no sense — neither scientifically, nor from a business perspective.”
After that, I decided to expand my horizons and look for opportunities beyond Montreal. When I visited Boston, it opened my eyes to the vibrant startup community that exists here, even at the time (early 2000s). I felt like a kid in a candy store in terms of the opportunity. While one company was working on completely new drugs targeted against disease-causing proteins, another was constructing DNA encoded library (DEL) for high-throughput screens, and yet another was developing novel next-generation sequencing technologies. There were just tons of new and cool things happening at the interface of technology and biology. I really felt that this [Boston biotech] was where I really needed to be.
During this time of exploration, I ended up meeting Stan Lapidus, who was then the CEO of Helicos. He is an electrical engineer by training and has been at the forefront of cancer diagnostics since the beginning. He invented the modern version of the Pap smear, “thinprep,” which became a huge product. He also led Exact Biosciences, where he helped invent a now standard diagnostic test for colon cancer without having to get a colonoscopy.
I met him and I told him that I wanted to be at the interface of business and science. Luckily, he said that they needed someone like me, and that I didn’t need an MBA. He gave me 48 hours to respond to the job offer. At the time, I was still meandering, talking to people and not prepared for an offer. But, I took the role as a Business Development Associate as my first job in industry. I jumped into Helicos, which was developing the world's first single-molecule DNA sequencing technology. I was finally working at the cusp of human knowledge and developing something totally novel. [At Helicos] it wasn't a lab job, but nor was it a “suit and tie” role. I was trying to figure out how to get this thing off the ground, build it up, and find value.
By the way, it wasn't a fancy job. Five or 10 years later, Stan [Lapidus] jokingly said my first job was carrying his bags. I was doing preliminary KOL [key opinion leader] outreach— getting into the offices of genomics leaders to expose them to this new tech and get input on the science. We wanted to better understand what we would build, how we would build it and what applications we could envision.
I was also doing a lot of competitive intelligence and looking at intellectual property [related to sequencing tech]. Then, I decided to take it upon myself to be the project manager for a behemoth of an instrument project. We were trying to build this machine that was arguably more complicated than an aircraft carrier. It had so many different parts and we needed multiple scientific disciplines to get it to work — optics, robotics, software, hardware, electronics, microfluidics, molecular biology and chemistry. Ultimately, we had to consider UI/UX as well and other productization requirements such as reagent stability, packaging, user guides etc.. This role was familiar to me in that my dad was a project manager for Bombardier, overseeing the construction of airplanes, which he originally would track manually with Gantt charts.
I was also given the role of product manager, responsible for applications, reagents, and defining what it was we wanted to achieve. I had no direct reports. I learned a ton about connecting science to value and fundamental scientific performance metrics — how do these relate to actual applications, markets, and customers? I learned how to not just analyze science, but also connect the dots to map new applications and markets. We had to synthesize multiple hypotheses to ask: What if we can do this for that purpose, for this market, or for these customers? What would be the technical requirements we would need to achieve? And would that be a compelling direction?
We had to deal with competitive headwinds from Illumina alongside economic downturns. Helicos went through a major downsizing, and I was one of the last people there. We had to reimagine the company. While we succeeded in building this cutting-edge single-molecule DNA sequencer, we failed to capture the market. Helicos had really high-powered VCs [backing us], including Flagship. At this point, I got a lot more exposure to how Flagship works. We started to ideate about how we could apply the technology to other ventures, rather than say: “Let's close up shop and move on.” Flagship has much more of an entrepreneurial or innovator’s mentality. [Together] we said: Let's apply the technology to diagnostics and imagine other applications.
4. What do you enjoy most about your current role at Flagship? What does it take to be effective in venture creation, specifically the Flagship model?
I think it is worth describing what we do at Flagship, because it will give you a better sense of why I love being here.
Our mission is to conceive of and then create first in category bioplatform companies. The words are important, because there's a real distinction between just creation of a venture and actually conceiving of it. Creation of a company could be interpreted as just putting people together, getting some IP and financing it. Conceiving of a company involves formulating the scientific nucleus, inventing and developing the technology, envisioning what it will be useful for, and serving as scientific entrepreneurs at the core of the whole venture. That’s what we do. We have permanent teams here—these aren't rotating roles nor is it the revolving door of an accelerator or incubator. We are permanent teams dedicated to innovating on a repeated and systematic basis. I lead a team, and there are five teams like mine, which ideate and invent. We then test [our ideas] in the lab, and run the experiments by hiring small wet lab teams to generate preliminary data that could validate those prospective ventures.
If the ideas pass muster, we build the companies around them. We will lead those entities for the first year or two, to get them off the ground. We build the team to a point where the company is ready to be more autonomous. We take the long view with these companies and finance them entirely ourselves initially, before inviting others to invest alongside us. We play a role that goes beyond simply oversight. We are very involved in strategy and implementation—the degree [of involvement] depends on the company.
This context is important because it harkens back to my itch to always be at the cutting-edge—to be a part of building a yet unmade future and always doing the most difficult thing. At Flagship you are perpetually doing that. It is what makes [working here] so compelling and exciting.
5. Which broad areas of science and medicine are you most excited about seeing develop in the next 5-10 yrs?
If you look at the portfolio of things that I'm working on it is quite broad. I wouldn't say there's just one or two things [I am excited about]. Just to name a few, one is the convergence of computation and biology. I'm super excited about that in part because of my adolescent interests [Q#1]. The digitization of large biological and clinical datasets and current computational power have enabled powerful new tools and algorithms. We are excited about that across the across the firm.
One of the first companies we started in this area was Cellarity. It is a pioneer in the field. We are still super excited about what it's doing, and what it can do. It is almost magical how computers and algorithms are much better at seeing things in the data than we are. I came from a lab where I was working on one protein, to now being agnostic to the proteins and their names and just letting the data and algorithms tell us what matters. This is one area that I think will continue to bear fruit. There are now quite a few companies applying AI to biology, but everyone's doing it in a slightly different way. Some are just focusing on applying AI to existing processes. I think there is plenty of space to increase the productivity and the likelihood of [drug discovery] success in biopharma.
Another company called Generate that we started in this area, but with totally different technology, focuses on using computation to create new biologic drugs—this is instead of using empirical, random [screens] and hoping something sticks. We are now in this regime of what we call generative biology, the ability to have data inform what the next novel design should be. It's not just understanding biology, [computation] is now generating biology, which is really cool.
I think new modalities are always going to be interesting. Whether it is a gene writer, gene editor, ASO- or RNA-based technology, all of these tools require a means of safely getting into the body and finding cells of interest. It’s funny that “delivery” used to be a bad word about five to seven years ago. Now, this concept is back in vogue and is going to be a critical aspect of anything we do. It's an exciting time to go beyond small molecules and biologics, and start designing smarter, more complex biological tools to cure disease.
What’s cool about a lot of the nucleic acid-based approaches is that we can use computation to rapidly prototype. We can now program these modalities and make them much more readily, which Moderna’s COVID vaccine demonstrated. We can go from the knowledge of a viral sequence to a drug or a vaccine, much more quickly than we could have done in the past. This is in part due to the “programmability” of these emerging platforms. This will help us respond to new and emerging threats as well, not just disease.
6. What is a non-Flagship biotech that you think serves as a paragon of both an impactful company scientifically, and has also created value for investors? What are the key learnings from this as a case study?
An obvious one perhaps is Alnylam. It is an impactful, programmable bioplatform company. It was a long journey, and there were many skeptics in the beginning. First, it was labeled as “science fiction.” Then, it became science debate and controversy. And then finally, after a long journey, it evolved into a company with multiple products that are that are impacting humanity. I think that it is exemplary of how I feel the biotechnology industry can have an impact. Hopefully the time horizons are going to get shorter—from vision to idea to product. Alnylam is a great example of people believing in a technology and realizing that we don't have all the answers today. We must work at it and put resources towards progress.
Again, we [in biotech] are creating an unmade future. We are not exploiting existing knowledge. If you're going to do science [create something new], you have to further the existing knowledge base, develop new techniques, and form new understanding of how biology works. You have to create new entities and see how they fare, and what their properties are. To me, this is the most exciting version of biotechnology.
7. What is one piece of advice you would give to new/aspiring biotech entrepreneurs?
There is a tendency in the newly minted PhD world to try to morph into a “business person”—and almost shed one’s scientific background. Perhaps this comes from an insecurity about being too “sciency.” If you are trying to build something totally new, and really be a pioneer, then science is the most important thing. Connecting science to value is the business side of being an entrepreneur.
Pioneering is exceptionally hard and requires you to do deep science. Do not simply persist and insist on your idea. Of course, you need to persevere and, at times, be somewhat stubborn but, at the same time, you have to be very scientific. You have to remain curious, inquisitive and evolve your idea with the data.
8. If you were in med school/grad school/business school today, and could sign up for a class, which topic/area would you choose? Why?
I would say, a course on computer science and machine learning. I actually took Andrew Ng’s class at Stanford, remotely, a couple of years ago. He teaches an introduction to machine learning course. I thought it was going to be a survey of topics but turned out to actually be a series of programming assignments. I had to stay up until midnight, trying to code and debug my algorithms.
9. What is a book you are reading in your free time?
Right now, I am reading Lee Kuan Yew’s memoir on the founding of Singapore. I'm interested in this for a couple of reasons. The first is that it feels like what one must do to create a startup, in some ways, which I find interesting. I'm also Armenian, and Armenia is a fledgling nation. I'm trying to understand the formation of Singapore and how it became so successful. There are parallels to Armenia, which is now trying to chart its future as a small, landlocked nation. There are similarities and dissimilarities, and I'm just trying to identify what those might be.
On the Foundation for Armenian Science and Technology and Flagship’s program:
There’s an institute, the Foundation for Armenian Science and Technology [FAST], which Noubar co-founded in Armenia.
FAST is trying to spur innovation, and entrepreneurialism in Armenia, particularly in the fields of computer science, technology and biotechnology. They have programs where they attract young graduate students and entrepreneurs from Armenia, who can then get training with us [Flagship] to learn about innovation. I've gone there and worked with the team in Armenia. In fact, a company has come out of this work. Armenian students have also come to Boston and ideated with us in a program akin to our 12-week Summer Fellowship Program. We believe that our innovation process is quite transferable [to Armenian students]. It is our hypothesis that conceiving of and creating companies or platforms is a “communicable addiction.” I think the program is bearing fruit. This work is really a passion of mine, and hopefully I'll be going to Armenia next month.