Many of today’s most trusted drugs are blunt tools. When successful, they bind to a specific target in the body to relieve pain or depression, stop a cancer, lower cholesterol, or suppress a cough. But they don’t stop there. They also bind to many other receptors, causing side effects that can range from a harmless dry mouth to life-threatening reactions.

These multi-receptor drugs are so notorious, they are known informally in the pharmacology business as “dirty drugs.”

Octant Bio recently raised a $30M Series A to do something about this. The Bay Area, CA-based synthetic biology startup is using an approach you may never have heard of, but that you definitely haven’t heard the last of.

Polypharmacology: A new approach

“Polypharmacology isn’t a term we made up,” says Octant co-founder and CEO Sri Kosuri. He says that many of the dirty drugs on which we depend work precisely because they bind to so many different receptors, often different aspects of the condition being treated. Kosuri’s team is leaning into the multi-receptor approach by applying advanced biotechnology in the pursuit of polypharmacology—the design of drugs that act on multiple targets or disease pathways. In doing so, Kosuri believes Octant can find not only drugs without unwanted side effects, but also better candidates for treating highly complex, multi-target diseases like obesity and diabetes.

“The last 30 years have been focused on specificity in the pharmaceutical industry,” says Kosuri. Using technologies like protein structure modeling, gene therapies, and CRISPR, drugs have been designed to hone in on specific targets: one gene responsible for a disease, or one biomarker that differentiates a cancer cell. “There have been a lot of successes,” he explains, “but most of these are in areas like rare disease and cancer, diseases that are genetically distinct in their nature.”

But for more common diseases with multiple causes, like cardiovascular disease, diabetes, Alzheimer’s, and mental illness, Octant speculates that this traditional thesis of drug delivery—find one target, drug one target—might not be the best approach. These complex diseases, which may emerge from the interactions of thousands of biological pathways, have proven elusive to companies still following the single-target trend. 

“The efforts to make today’s dirty drugs more specific haven’t led to more effective treatments,” says Kosuri, “not because we haven’t been able to make drugs more specific, but because those drugs simply didn’t work as well.” This may be one reason why drug innovation has been sluggish for complex yet common diseases.

The special sauce in Octant’s technology 

In the past decade, two complementary fields have emerged from academic labs to revolutionize the life sciences industry: synthetic biology and computational biology. Synthetic biology uses genetic tools and techniques to design biological systems, while computational biology uses the growing power of machine learning (a.k.a. artificial intelligence) to analyze massively complex biological data. While these two toolsets are powerful on their own, combining them can result in unlocking completely new tools and platforms for the life sciences and drug discovery. 

Octant combines synthetic biology and computational biology to create a platform that can measure the impact of a drug across cellular pathways. Octant’s early focus is on a group of cell signaling receptors called G protein-coupled receptors, or GPCRs, which live on the outer surface of human cells and trigger activity within the cell. (About one-third of all drugs are GPCR modulators, accounting for around 10% of global pharmaceutical sales—around $100 billion.) 

Octant engineers cells so that when a drug affects one of these GPCRs, the cell releases a unique, genetically encoded barcode signal. By measuring these signals across thousands of engineered cells, Octant maps the full effects of a chemical across hundreds of potential GPCRs. By analyzing the data from these high-throughput experiments, Octant can identify the optimal chemical to hit a desired set of targets in just the right way.

Octant uses this platform to see how current drugs, failed drug candidates, and natural chemicals hit the hundreds of GPCRs in a given human cell. By comparing a chemical’s “hits” to its known effects and toxicities, Octant can then predict a set of targets that will both treat the disease and avoid unwanted side-effects. 

“Octant is built around two goals,” Kosuri said. “The first is identifying the polypharmacological sweet spots we might want to hit, and second is how we build small molecules to do that.”

After identifying those sweet spots, Octant turns to its synthetic biology platform, using high-throughput experimentation to test a massive array of chemicals, and then refining to find the best molecule for hitting a particular target set. In contrast to the hypothesis-driven target identification process, Octant’s asset discovery program is empirically-driven. It embraces high-throughput experimentation, big data, and machine-learning similar to that of companies specializing in computational biology (for example, Recursion Pharmaceuticals or AbCellera). This is the beauty of Octant Bio’s platform—they’ve built a way to collect biochemical data en masse, and then analyze it both rationally with human researchers and empirically using computational engines.

Octant’s platform has the ability to detect small numbers of RNA molecules. In response to Covid-19 the company repurposed this feature for use in detecting the virus at the heart of the pandemic. Octant made this method, called SwabSeq, freely available to the research community to help overcome certain bottlenecks in testing for coronavirus, including RNA purification, qPCR machinery, and automation. Octant reports that a number of groups, firms, and academic institutions have already utilized the SwabSeq protocol.

The bigger picture: An atlas for treating complex disease

Octant’s big-picture vision is not limited to the discovery of individual drug candidates. The company also aims to build a definitive map of chemical-cell interactions, with the ultimate goal of providing a chemical solution to any complex human physiology. If you can find a set of GPCR target vectors associated with a disease, Octant can design the chemical. (The platform can even be applied to flavors and fragrances, which our bodies sense through taste and olfactory GPCR receptors.)

In the shift to a data-first, computational biology approach, Octant embraces a business model vastly different from the high-risk, high-investment model seen in traditional pharma. By acting as a central resource which any researcher can use to find and build a multi-target drug candidate, Octant can 1) diversify its pipeline, 2) minimize the in-house time and cost of identifying disease target vectors, and 3) focus on diseases that have known target vectors. With their unique chemical-to-cell mapping, Octant will be able to make the most targeted polypharmacological drugs, and they will be able to do so more quickly and cheaply than potential competitors.

A different kind of team, a different kind of strategy

Octant has a unique pedigree. Kosuri himself is the former staff scientist of George Church, a synthetic biology pioneer and serial entrepreneur. Kosuri’s co-founder is Ramsay Homsany, formerly of Google and Dropbox, who serves as President. Octant’s board of directors notably includes Jason Kelly of Ginkgo Bioworks (synthetic biology’s first unicorn), the academic luminary Charles Zuker, and Vijay Pande, a Stanford biophysicist and partner at Andreessen Horowitz, which anchored the company’s Series A. What does this team see in Octant’s approach?

“There are a lot of people working on cancer immunotherapy and new ways to achieve specificity through AI,” Kosuri tells me. “But no one else is working on the types of things we’re working on.” At a time when pharma companies are avoiding the messy search for complex disease treatments, Octant and its leaders believe that doubling down on complexity is going to lead to the future of drug discovery.

“We’re making a fundamental bet that we’re not going to find a single target,” he says.

Octant’s real bet is that better tools for understanding the basic biology behind complex diseases will soon be developed. These tools will enable researchers to identify new and promising potential drug treatments. Such discoveries may be made by Octant itself, by a cousin firm, or one of the many academic labs and biotechs developing disease-specific research tools and models. Whoever the discoverer may be, Octant will be their natural partner for translating basic science into treatments and cures.

Follow me on Twitter at @johncumbers and @synbiobeta. Subscribe to my weekly newsletters on synthetic biology. Thank you to Matthew Kirshner for additional research and reporting in this article. I’m the founder of SynBioBeta, and some of the companies that I write about—including Ginkgo Bioworks—are sponsors of the SynBioBeta conference and weekly digesthere’s the full list of SynBioBeta sponsors. I am also an operating partner in venture firm DCVC, which has invested in Recursion Pharmaceuticals and AbCellera.