Editor’s Note: This post first appeared as an issue of the a16z Bio Newsletter. To receive this newsletter, please sign up here.
In this post:
The vast majority of healthcare spend, operational transactions, and encounters occur through traditional clinics, hospitals, and insurance companies—all entities it’s been very hard to sell tech to in the past, as the first cohort of health tech companies in the early 2010’s learned painfully. Because of the tech immaturity of those buyers at the time, they were largely unable to absorb startups’ solutions quickly enough to support high-growth, venture scale business models.
In response, the next generation of health tech startups shifted to building around the system, in the form of direct-to-consumer services and full-stack companies (primary care providers, pharmacies, insurance companies, appointment marketplaces, etc). Consumers were so fed up with what was broken in the traditional healthcare system—lack of access, convenience, and affordability—that a group of early adopters were willing to pay out of pocket to new companies for lower wait times and better customer service. That pent up demand meant that many of these startups began to experience rapid early growth—and many continue to exhibit astounding early user acquisition and revenue metrics that are unprecedented in the digital health space. But the hard truth is that there will be a limit to this trajectory, and these companies will have to face the inevitable challenge of integrating into the broader supply chain of healthcare to continue to get to true scale, both in terms of the surface area of their products, as well as their reach into broader patient populations.
Meanwhile, providers’ and payors’ technology infrastructure has matured significantly, along with a broad recognition that EHRs and other core systems are not fit to solve today’s business challenges in delivering value-based care, care coordination, and improving customer experience. Incumbents are also at a tipping point of financial pressure, due to payment model reform and competition, compelling them to innovate in big and different ways. So we will soon see a third generation of health tech startups emerge, many as full-stack, consumer-centric companies who (thanks to that second generation!) have renewed courage to sell directly to and partner with the traditional players, with the reward of achieving massive scale and transforming the system through its core.
Bioengineering, once viewed primarily as an academic discipline, is growing up. Tools and treatments that are engineered, not discovered, are now making their way not just into new startups but into established industry and major biopharma companies—massive organizations built on the foundations of discovery, ingesting companies built with an engineering DNA.
Now come the culture clashes. In biopharma and healthcare today, the “old” culture of discovery — the idea that science is driven by discovering new knowledge (hypothesis —> test —> repeat) — is clashing with the “new” culture of engineering (design —> test —> iterate). This clash encompasses how everything is handled, from identifying biological targets to designing clinical trials and even to how we access health care. In this article published in STAT, I talk about the 4 major culture clashes we’ll see as these two worlds and mindsets increasingly intersect—driving us forward into the future. Welcome to the bioengineering culture clashes.
Tech and biotech just don’t mix. At least that’s the conventional wisdom. But the intersection of the worlds of biology, computer science, and engineering has created a new hybrid of tech + biotech that we simply call “bio”. World, meet bio. Bio, eat world.
In this new bio world, the well-worn playbooks are out of date. Bio is blurring lines and dissolving silos across the entire healthcare industry. In this post, I tackle 16 enduring myths, misconceptions, and sacred cows that still persist in traditional tech and biotech circles when it comes to bio (e.g., Scientific founders can’t be CEOs! Silicon Valley can’t do biotech!)—and why they’re the wrong mindsets for this new world.
From 2000-2010, we saw the birth of the conceptual framework for synthetic biology, in early “toy genetic circuits”—essentially, simple tools to turn proteins production in cells on or off. Many predicted that our ability to program in genetic code would create a waterfall of new products and revolutionize every industry. While people had used microbes to produce molecules for years, the first big, concerted effort to chase an application for synthetic biology was the production of biofuels, which didn’t turn out to be the best application (biofuels were too cheap of a product, scaling was too hard, etc). The biofuels bust in the early 2010’s caused many companies to pivot into making more valuable chemicals like flavors and fragrances. But the sad truth about most of those chemical production applications of engineering biology is that they actually don’t require the intricate complexity and full power that synthetic biology techniques are capable of: building cells that can sense, compute, even respond.
The last decade saw synthetic biology finally find a killer app for the complexity available to it: cell therapies. Some early ideas included bacteria engineered to kill tumors, or engineered cells that could sense and respond to sugar in the bloodstream to help manage diabetes—but these efforts were made too far removed from the clinical community to have a realistic path to market. In the meantime, however, Carl June’s lab was doing the groundbreaking work of expressing chimeric antigen receptors (CARs) in T-cells, which normalized the idea of engineered cell therapies. Following this breaking of the ice, groups like Cell Design Labs and others applied the mindset and tools of synthetic biology (switches, logic gates, etc) to these CAR T therapies. Now the path to engineered cells with dynamic sense-and-respond capabilities is much clearer.
CAR T therapy’s cancer killing ability was only the first killer app for synthetic biology tools and techniques. There are endless applications for advanced biological computation that once sounded far-fetched: patterned materials that can self-repair when they sense damage; highly parallel computation across a population of cells or molecules to outperform silicon; perhaps even an automated “cellular recorder” for what food, medicine, and exercise a patient has experienced. As we continue to see new generations of scientists and founders trained in the mindset of “engineering biology”, we will see more and more clinical communities and other industries embrace these new tools, leading more and more possibilities and many new applications. The biggest obstacle is no longer technical, it’s finding the right applications where the market opportunity can justify high development costs.
History will remember the past decade as the coming of age moment for cell and gene therapies. A field once mired with setbacks has closed off the decade with 4 FDA approved medicines (including 2 CAR T drugs and 2 in-vivo gene therapies) and its most groundbreaking tool, CRISPR, make its way to clinical trials in humans.
This coming decade will be the golden age of translating these “living medicines” into practice. As cell and gene therapies make their way into part of our standard therapeutic armement, we will begin to see their full-scale industrialization. AI and automation will transform the laborious and bespoke elements of design (genome and cell engineering), manufacturing (e.g., vectors and cells) and delivery of these therapies (supply chain and logistics) into much more efficient processes. The cell and gene-editing “developer community” will continue to expand our collective toolkit, allowing us to perturb and engineer biology in new dimensions and at levels of precision that were previously unscalable. New protein engineering, genetic circuits, and delivery system innovations will change the paradigm of how we use and dose these medicines.
As the infrastructure to produce these drugs matures, cell and gene therapies will start to go mainstream: treating many more chronic diseases and conditions that the vast majority of the population experience—cardiovascular and neurodegenerative disease, solid tumors, maybe even aging. On the flip side, we will also see the door open to “n of 1”, precision medicines for single patients with very specific genetic ailments (once impossible from a technological and financial standpoint). Beyond pure drugs, this technology is also ripe to help us realize many of the long promised innovations in regenerative medicine—organ replacement, tissue regeneration, and even engineering stem cells to be the foundation of new off-the-shelf cell therapies.