Get this monthly update from the a16z Bio Team: sign up here.
Welcome to the first edition of the a16z Bio Newsletter, where we share, once a month, what we’re seeing at the intersection of engineering and biology — from news and trends to thoughts on how these affect not just healthcare, but all industries.
Here’s the big shift: Biology is going from an experimental science to an engineered discipline. In medicine, this is already driving an explosion of new modalities and engineered “living medicines” (in the form of genes, microbes, even applications). And because these medicines are modular and programmable, each successive generation of therapies can be iterated and built on — just as with other engineered applications — to treat other diseases. Of course, none of this will be easy, without risk, or cheap — but it does fundamentally change how we diagnose (Dx), treat (Rx), and even prevent (Px) disease.
A new generation of care delivery companies
As technology changes how we access, pay for, and deliver healthcare, we’re seeing new types of companies as well as the entire industry structure re-aligning. Just look at all the recent M&A (Walmart and Humana; Aetna and CVS; Amazon, JP Morgan, and Berkshire Hathaway), which is creating new “hybrid” companies that straddle broad swaths of the healthcare value chain. So where does software come in? It has the potential to both remove friction within the system — including blurring traditional silos — and to dramatically expand the ways we access care. Companies with technology built in at their core — whether tech incumbents entering the healthcare arena or those purpose-built for this from the ground up — are leading the charge.
Biology is eating the world
Biology — with its unparalleled ability to evolve, replicate, and create — is one of the most advanced manufacturing technologies on earth. This is why we will see biology increasingly become a part of every industry. We’ve already seen it transform food, agriculture, textiles, manufacturing, and with DNA-based computers, even software… what’s next? Much like the last century was transformed by information technology, we are now living in the century of biology. In short, bio is eating the world.
A new era of damage-free genome editing with the next phase of CRISPR
The first wave of CRISPR (essentially a DNA cutting tool) represented the possibility of precise, one-and-done, gene therapy cures that could rewrite the genetic errors behind some of the most devastating diseases. Now, we are entering a second wave of innovation in this space — going from a cutting widget to a full-fledged toolkit for medicine (as well as better Cas systems) — that expand the power of CRISPR across therapeutics, R&D, and diagnostics.
New molecular technologies show how we can build on top of CRISPR as a platform with a constantly growing suite of programmable functionalities. Just last month, teams under scientists Samuel Sternberg and Feng Zhang both published papers showing that a CRISPR-Cas system paired with bacterial transposons can be programmed to insert new DNA into precise locations in the genome without cutting the DNA strands. The researchers came to this insight by noticing that the transposons — aka “jumping genes” — seemed to jump around the genome aimlessly (and who knows why, perhaps evolution?), but did so through some RNA-guided mechanism, which is not unlike how CRISPR systems are programmed.
Getting rid of the need to cut the genome — which conventional CRISPR-based DNA editing depends on — avoids the mutation-prone DNA damage response, which can trigger genomic deletions, translocations, or other collateral damage. Additionally, this technique could efficiently integrate very long strands of DNA, potentially enabling applications that require integrating large genetic payloads.
Such innovations in precise, cutting-free genetic modification push us one step closer to the holy grail of predictable, highly efficient, and damage-free genome engineering. The steady flow of new technologies like this — built by the genome engineering “developer community” — is what will allow the field of biology to be truly transformed by an engineering approach, unlocking new medicines. –Andy Tran, bio deal team
Bonus Content: Check out the a16z Journal Club audio podcast discussion of these papers, with general partner Jorge Conde and Bio deal team partner Andy Tran
The birth of the digital health formulary
Recently, Express Scripts and CVS announced “Digital Health Formularies”: curated lists of healthcare apps and programs to help payors and consumers make better informed decisions about what apps to use, and under what circumstances. With now 100Ks of digital health apps available to consumers, those responsible for patient outcomes and footing the bill are increasingly concerned about appropriate utilization in this space, which is where formularies come in.
It’s the first time digital health is being treated in much the same way as traditional medications, and signals a tipping point in the maturity curve for the digital health category. That’s because formularies help payors make sense of their options as the number of treatment options and rates of utilization rise, from providing guidance to patients to managing costs in a given therapeutic or service category.
It makes sense for PBMs (pharmacy benefit managers) like Express Scripts and CVS to step into this role, but we’ll likely see several players, if not an entire ecosystem, emerge here, too. For example, companies like Accolade help consumers navigate, with personalized guidance, on what apps to use; other companies like Xealth help providers embed digital health formularies into their point-of-care workflows, literally allowing apps to be prescribed to patients from a pre-vetted menu.
But digital health formularies will require new criteria to be considered — such as data security and privacy, usability, interoperability, and accessibility. While stakeholders will have to be mindful of administrative overhead that other benefit management plays have created historically, the introduction of such marketplace-like dynamics will help separate the wheat from the chaff among digital health players and apps.
The ability to make an organism create entirely new proteins
In a series of scientific papers in the late 1990s and early 2000s, Peter Schultz and colleagues set up a grand challenge for biologists: to adapt the genetic code to enable it to produce new proteins that can do specific tasks no natural organism can accomplish today (Liu & Schultz, 1999, PNAS). And last month, researchers at the MRC Laboratory of Molecular Biology in Cambridge, UK, led by Jason Chin, made a big leap toward this goal with a paper titled (somewhat hyperbolically!) “Total synthesis of Escherichia coli with a recoded genome”. The aim of this work was to create an E. coli cell that could make those never-before-seen proteins using an expanded genetic code.
This represents quite a feat of engineering biology. Many interdependent parts in the machinery of E. coli rely on a common genetic code to translate between the language of DNA (with 64 ‘words,’ a.k.a. codons) and the language of proteins (only 20 ‘words’, a.k.a. amino acids). Cells have translator molecules called tRNAs that can map DNA ‘words’ to protein ‘words’, linking a codon to the right amino acid. But because the mapping isn’t 1:1, a single ‘word’ in the language of proteins may have up to six ‘synonyms’ in DNA, meaning some sets of DNA sequences are redundant.
Scientists at the MRC took advantage of this redundancy to create new protein ‘words’ (amino acids) that they could string together into proteins. To free up three DNA ‘words’ (codons), they changed all instances of these codons to their synonyms. Then they modified the translation machinery to map those codons to new, unnatural amino acids. Since all instances of the original codon had been removed, this incorrect mapping didn’t affect the cell health. Now that they have an organism with an expanded genetic code, they can use the newly mapped codon to incorporate unnatural amino acids into any protein they choose to produce. Sounds far easier in theory than it is in practice: changing these three codons to their synonyms required modifications at 18,214 positions in the genome, all without affecting the cell’s ability to grow. To do this they had to chemically synthesize and assemble the entire genome: four million base pairs!
No one can say yet what the best application for these recoded organisms will be, whether in medicine or manufacturing or beyond — but the ability to make an organism create entirely new proteins will certainly be a supertool of some sort for biological engineers. Some startups have used recoded organisms to develop protein drugs (biologics) with increased stability and slower degradation; similar strategies could be used to adjust medicinal properties in biologics, like altered targeting or controllable degradation. You could even imagine using organisms like these as hosts for producing unnatural enzymes and for performing new types of reactions to manufacture high-value molecules. In short, the development of a new and valuable tool of the biological engineering toolbox might be on the way.
Why multimillion dollar drugs are here to stay
This past month, we saw the approval of not one but two gene therapies with approximately $2M price tags… each. Bringing us into a new era for drug prices, this news led to heated discussions in media, policy, and healthcare about how we might value and pay for these drugs.
Both treatments deliver genetic material into patient cells to treat disease, and are some of the very first to be approved in a category of one-time treatment drugs that replace lifetimes of arduous chronic therapy for patients: Zolgensma, a gene therapy approved to treat children with spinal muscular atrophy (SMA), a rare disease that is a leading genetic cause of infant mortality; and Zynteglo, a therapy for a rare blood disease.
Precise, one-and-done, curative therapies: these medicines bring us significantly closer to the holy grail of medicine. But how do we pay for these drugs? How can we place an economic value on what they are worth? And how might we change policy — with one slash of a pen — so more patients can have these life-changing medicines paid for? I deep dive on this topic with esteemed MIT economist Andrew Lo to discuss a novel approach, here.