When Marc Andreessen and Ben Horowitz co-founded the Silicon Valley venture capital firm bearing their surnames in 2009, the general partners initially avoided investing in life sciences or healthcare. They reasoned that these sectors were too complex to dabble in and would not lend themselves to impactful technology along the lines of their normal investment sectors.

Six years later, however, the rise of artificial intelligence (AI), genome editing, and other technologies—and their adoption in biotech and healthcare—convinced Andreessen Horowitz to expand into those sectors. A great deal of credit belongs to the person hired to oversee that effort—the firm’s ninth general partner Vijay Pande, PhD.

Vijay Pande, PhD, a general partner at Andreessen Horowitz (a16z) and the founding investor of the firm’s Bio + Health fund.

Pande, the Bio + Health Fund’s founding investor, launched Andreessen Horowitz’s first biotech/healthcare fund in 2015, the $200-million AH Bio Fund I. That was followed by three additional funds, the latest being the $1.5-billion AH Bio Fund IV last year. Combined, the four funds total $3.2 billion—about 10% of the total $35 billion in assets under management of Andreessen Horowitz or “a16z” (The nickname denotes 16 letters in the firm name beginning with “a” and ending with “z”)

Pande leads a team of about 40 professionals who oversee the Fund’s biotech and healthcare investments, which a16z describes as being at the cross-section of biology and computer science, including applications in computation, machine learning, and AI in healthcare; digital therapeutics; diagnostics; and other novel transformative scientific advances applied to industry that take bio beyond healthcare. Depending how healthy the investment climate proves, that headcount could grow this year to 50.

The Fund’s investment thesis asserts that biology has shifted from empirical science to an engineering discipline—from using human-created approaches for controlling or manipulating biology, to applying using nature’s own machinery through biological engineering to design, scale, and transform biology.

a16z lists more than 50 portfolio companies for its Bio + Health Fund, of which 24 are biotech focused (including six in synthetic biology). They include:

  • insitro, the machine learning-based drug discoverer and developer led by founder and CEO Daphne Koller, PhD. The company, which raised $400 million in series C financing in 2021, combines induced pluripotent stem cells (iPSCs), genome editing, high-content cellular phenotyping, machine learning, and other data-generating tools to build in vitro models of disease designed to maximally predict human clinical outcomes.
  • Scribe Therapeutics, which develops CRISPR-based treatments through its genetic modification platform. Co-founded by Nobel laureate and CRISPR pioneer Jennifer Doudna, PhD, Scribe raised $100 million in series B financing in 2021.
  • Tmunity, which applies a multiplatform approach to build an innovative “toolbox” of orthogonal assays to provide a comprehensive set of data to characterize CAR-T cell therapeutic products. a16z led the $75-million series B round of Tmunity, whose co-founders include cell therapy pioneer Carl June, MD.

In an exclusive interview with GEN Edge, which is being presented in two parts, Pande discusses a16z’s expansion into engineered biology and healthcare, how differently biotech and healthcare have embraced the engineering paradigm, and the potential for engineering biology to cut drug development costs.

Pande was previously the Henry Dreyfus professor of chemistry and professor of structural biology and of computer science at Stanford University, and concurrently served as director of the biophysics program at Stanford. While there, Pande founded the Folding@Home Distributed Computing Project for disease research, which pushed the boundaries of computer science techniques (distributed systems, machine learning, and exotic computer architectures) into biology and medicine, in both research as well as the development of new therapeutics.

(This interview has been lightly edited for length and clarity; Part II can be read here.)

GEN Edge: How did you convince Andreessen Horowitz’s co-founders to establish the a16z Bio + Heath Fund?

Vijay Pande: I was able to convince them, originally in 2014, that there was a major inflection point, and that tech is really changing how life science is done. When you think about computation, AI, or even beyond computer tech, you’re thinking about an engineering mindset, like gene editing, cellular therapies, mRNA vaccines. That mindset is having a big impact on the life sciences.

We’re seeing something very analogous on the healthcare side, where again there can be computation like machine learning impacting how healthcare delivery is done. But also a tech-like mindset, almost an Amazon-like mindset, where there’s not a killer algorithm, but a sort-of tech applied to logistics that I think is really transforming how healthcare delivery is performed.

With that in mind, they actually turned around 180° and said, “Let’s actually build a huge effort and build out the first new vertical within the firm.” With that, we launched the first fund. Now, we’re investing our fourth fund, a $1.5-billion vehicle!

GEN Edge: a16z says the Bio Fund is committed to financing entrepreneurs and their startups focused on engineering biology and reshaping healthcare. How does a16z define “engineering biology”?

Pande: When you think about engineering, there are different forms. One form is just design over discovery. A lot of basic research has to be done with discovery, but we don’t discover bridges—we design bridges! If you think about various areas, there are actually new ways of developing small molecules, cellular therapies, or mRNA vaccines—all of those things are really designed instead of discovered.

One of the key elements of that is a sense of reusability—when you develop a cellular therapy, you have to come up with a new guide RNA, but you actually can reuse almost all of the rest of the machinery. Similarly, with mRNA vaccines, the process of making new vaccines is very much automatable. The time to go to clinic is therefore much shorter. And also hopefully, the chance of success is much greater. A beautiful example of this is the COVID-19 vaccines, which are themselves engineered vaccines.

GEN Edge: How does that engineering perspective shape the fund’s approach to investing in biotech and healthcare?

Pande: What that usually means is that we have not emphasized single-asset types of life science companies. We want to invest in platforms, where the company builds the ability to engineer something, whether we’re talking about engineering a cancer diagnostic like Freenome, engineering small molecules like insitro or Genesis Therapeutics, engineering CRISPR enzymes like Scribe Therapeutics, and so on. Or maybe it’s engineering how healthcare is done, like a company like [healthcare AI platform developer] Bayesian Health.

In all those cases, design is one level of engineering. There’s even a stronger level, which is design-test-iterate, and the ability to improve is actually a hallmark of engineering. In fact, things like Moore’s Law or Flatley’s Law in genomics, the fact that these systems improve exponentially year-over-year, decreasing cost, increasing capabilities—that’s something that just comes if you can, let’s say, get 15% better every year, you get exponential benefits.

GEN Edge: When it comes to adopting and embracing engineering technology, how would you compare where biotech is vs. healthcare?

Pande: It’s hard to compare in terms of who’s ahead, but they’re doing it in different ways. In life sciences, if you can develop a therapeutic with an engineering approach, and it does well with trials, how it got there isn’t as important to the patient as much as the fact that it works. It doesn’t matter to the doctor.

In healthcare, you’re going to have to work within the healthcare system, and changing how doctors work is therefore a key part of this. And one of the ways that we’re seeing companies have a big impact is to really get into the flow of how healthcare is done. One of the most dramatic examples is probably Devoted Health, which is a payer and a provider. And they use technology in all parts of that stack. So, they can impact care, because they’re the ones performing the care.

GEN Edge: Yet healthcare is generally thought to be a little bit behind the curve compared to life sciences in embracing engineering tech.

Pande: I think that’s probably right in some ways, but it might be catching up very quickly. Healthcare has been slower to adopt technology, definitely, but I think that is rapidly changing. It’s clear that healthcare cost is a huge crisis for our country. It’s soon to be 25% of U.S. GDP, and growing. We have to do something to address those costs, and I think people are looking to innovation and technology as a very natural next step.

GEN Edge: If there’s an issue of engineering and design, what does that do for the therapeutic startup or early-stage company? How does engineering make a difference in that world?

Pande: What you can do is a couple of things. One is that you could actually get to your drug candidate faster. You could hopefully design better drug candidates that have greater efficacy, less off-target affinity, and so on. But also, a lot of this engineering mindset goes into even identifying the right targets, and using machine learning and artificial intelligence to use human data, whether it’s human cell lines or human clinical data, to be able to predict the best targets.

In the end, we all know mice are pretty poor models of humans, and so much of the challenge that we have in clinical trials comes from understanding mouse biology and not understanding human biology. Here, AI allows us to develop a model that, while flawed as all models are, AI being data driven could actually be a much better model of people than real mice are of people.

GEN Edge: Does that mean we can expect to see fewer mouse studies?

Pande: Yes, I think in time, definitely. I think, for better or worse, there will still be probably non-human primate studies that will be done. But one of the other interesting things is that AI can build upon previous learning. So, as this data gets done, hopefully, we will start to eventually get to a critical mass of data, such that less will be needed.

GEN Edge: In one of your commentaries, you listed principles of engineering biology that included Lego-like building blocks, repeatability and reproducibility, testing, and process engineering. To what extent will this reprocessing into engineering make drug development cheaper?

Pande: One way that it makes it cheaper is that for any of these methods, you can reuse many elements. We see this in gene editing, and I think we’ll see this in mRNA—the editing elements, the delivery elements. Then the only part that differs might be, let’s say, the guide RNA. In that sense, it will require a lot less money to be able to get to the clinic. Hopefully, as we’re driving this with human biology, we’ll have a much better chance of success in the clinic. And when you have fewer failures, it means then that the overall cost goes way down.

One other huge opportunity is using AI to design clinical trials. And that could be patient stratification, patient identification. In that sense, the dream of personalized medicine may actually start to become realized, and driven by cost, driven to make much more successful clinical trials, and then be thinking of patient selection from the very beginning.

GEN Edge: Was cost the sole or main driver of this move into engineering? Or are there other factors?

Pande: It’s an interesting thing: All universities have bioengineering departments. Those departments are maybe 10 to 15 years old. So it’s a relatively new movement, and the concept of engineering biology itself is relatively new. I think it’s that sort-of rise as an intellectual discipline that has really made an impact now in translational medicine.

GEN Edge: Where do modalities like genome editing and gene therapy and even newer ones like base and prime editing fit into this paradigm?

Pande: These are just different tools in the engineer’s toolbox. We can talk about these as tools for actual therapeutics, but they’re heavily used right now ex vivo for research, and to decode human biology as well. These things will start to have multiple impacts where you’ll be able to understand biology better as you can perturb biology.

GEN Edge: Another area of apparent interest is aging, given its role in triggering Alzheimer’s disease. Lately, we’ve seen some companies pursue aging research in and of itself. How will this paradigm shift to engineering biology affect aging research?

Pande: One of our portfolio companies is a beautiful example—BioAge Labs. They use machine learning applied to aging datasets, where they have datasets from people who have lived 80, 90, 100 years. In some cases, they have longitudinal data on these patients, and blood samples with longitudinal data.

The first place where longevity is interesting is that the biology of aging is a relatively underappreciated and underleveraged part of understanding biology. There are many diseases—Alzheimer’s, also cancer, type 2 diabetes—that are greatly exacerbated with age. COVID-19 itself is an example too: kids don’t get nearly as devastating impacts from COVID-19 compared to the elderly. Understanding the biology of aging will suggest new targets, and those new targets will then allow for treatments for specific therapeutic areas. This is something where you don’t have to target aging itself, as much as learn from the biology of aging to target existing therapeutic areas, existing disease.

BioAge has a drug for muscle atrophy, which in principle could be given to patients that are on ventilators, where basically your lungs atrophy because you haven’t been using them. But in their Phase I readout, actually patients who are just sitting in bed, normally if you sit in bed for 10 days, you’ll lose a great deal of muscle. And they have patients who basically did not lose muscle, and some of them gained muscle, which was kind of shocking.

GEN Edge: How has COVID-19 helped advance engineering biology?

Pande: I think there are a couple of different areas where COVID-19 is advancing engineering biology. One is just the vaccines themselves being the first example of programmable and engineered medicines. It is very inspirational. Also, the fact that we could get things into patients rapidly, when there was the right sort of setup and situation to do it is very intriguing. That’s from a life sciences point of view.

Also from a healthcare delivery point of view, people are much more thoughtful about their healthcare. They are being the driver of it themselves, rather than just purely responding to doctors. I think there will be a huge surge in consumer-oriented healthcare, especially as healthcare insurance shifts to higher deductibles, and a mindset that the patient should be a part of this process rather than just the product itself.

GEN Edge: Any concern that consumer-driven healthcare thinking will lead to some consumers saying, “I don’t need meds, I don’t want to take a vaccine.” We’ve seen some of that in this country.

Pande: That’s going to be here whether there’s consumer-driven medicine at all. My doctor friends would say, “Oh that’s great. I just wish my patients would do what I told them: Just take the pills, do the workout, don’t eat too much,” and so on. Compliance, if gamified, could be a very interesting application of tech.

Previous articleThermo Fisher Scientific Expands Single-Use Technology in China
Next articleGenetic Risk Factor for AMD Identified by Combining Experimental and GWAS Data