Building drugs for the cancer targets everyone else called impossible.
In drug discovery, "undruggable" is usually a polite way of saying the industry gave up. Some of the most important proteins in cancer - transcription factors that switch cancer-driving genes on and off - have smooth surfaces with no obvious pocket for a small molecule to grab. Antares Therapeutics, a Boston company that launched in June 2025, treats that word less as a verdict and more as a list of jobs to finish.
Antares is a spin-out of Scorpion Therapeutics, formed to advance Scorpion's remaining pipeline after Scorpion sold its lead PI3K-alpha program to Eli Lilly in early 2025. Rather than wind the leftover assets down, the founders spun them into a new company with a single, unfashionable thesis: the targets nobody can drug are exactly where the unmet need - and the value - is concentrated. The name itself carries the lineage. Antares is the brightest star in the Scorpius constellation, known as "the heart of the Scorpion."
The company arrived with real money behind the idea. Its $177 million Series A was co-led by Omega Funds, Atlas Venture, Lightspeed Venture Partners, BVF Partners, and Cormorant Asset Management, with participation from Abingworth, Invus, Vida Ventures, Tenmile, Willett Advisors, and others. Within a year, a bet that had yet to put a single drug into a human being attracted a collaboration with Novartis worth up to roughly $1.9 billion. That sequence - big science first, big pharma validation second, clinic third - is the whole story in miniature.
Most successful small-molecule drugs work because the target protein has a natural cavity - an active site or pocket - into which a molecule can slot and interfere. A large fraction of biology's most consequential proteins simply don't. Transcription factors, certain scaffolding proteins, and a range of oncology targets have been "validated" for decades in the sense that scientists know they drive disease, yet they remained out of reach because there was nothing obvious to bind.
Antares' answer is to hunt for pockets that are transient, allosteric, or only appear when a protein moves - and, where possible, to form covalent bonds that lock a molecule in place. It combines proprietary compound libraries, next-generation mass spectrometry, and chemical proteomics to physically detect these fleeting binding sites, then uses physics-based modeling and machine learning to design molecules that exploit them. The result: candidate drugs aimed at proteins that have frustrated the field for a generation.
Antares doesn't sell one hero molecule. It sells an integrated engine - three ordinary disciplines wired together tightly enough to do what none could alone.
Proprietary compound libraries paired with next-generation mass spectrometry and chemical proteomics to discover brand-new chemical pockets on proteins - including covalent and allosteric binding sites.
Human genetics and functional genomics to validate which targets truly drive disease, prioritizing first-in-class or "best-through-first" opportunities with strong biological rationale.
Physics-based modeling, molecular dynamics, and machine learning to predict how molecules bind and to accelerate the leap from a fleeting pocket to a clinical candidate.
Antares funds its own discovery through venture capital, then monetizes the engine through partnerships: upfront payments, option-exercise fees, development and commercial milestones, and tiered royalties - while advancing wholly owned programs toward the clinic. The two headline numbers below tell that story.
Bars scaled to the $1.9B Novartis deal ceiling. Figures per company and press announcements, June 2025 - June 2026.
June 2026 collaboration to discover, develop, and commercialize first-in-class small molecules against undruggable oncology targets. Structure: $105M upfront, up to ~$1.9B in total option and milestone payments, plus tiered royalties reaching low double digits. Antares runs research and retains control until Novartis exercises its options.
A drug-discovery collaboration focused on transcription-factor targets - among the hardest and most sought-after territory in oncology.
Associated clinical-stage EGFR inhibitor assets (PFL-721, PFL-241) referenced within the company's broader program landscape.
Leads the company's thesis of systematically unlocking high-value, hard-to-drug targets.
Directs the science across chemistry, biology, and computation.
Guides clinical strategy as the lead program nears human trials.
Oversees the path from discovery into development.
Runs the medicinal-chemistry engine and compound libraries.
Leads the physics-based and machine-learning discovery work.
Board chaired by Jeff Albers, J.D., M.B.A., with Keith Flaherty, M.D., Jean-Francois Formela, M.D., Shelley Chu, M.D., Ph.D., and Sir Menelas Pangalos, Ph.D.
Many biotechs bet on one breakthrough technology. Antares wires chemistry, biology, and computation together so tightly that the whole finds pockets none of the parts could - a systems bet, not a single-tool bet.
Next-generation mass spectrometry and chemical proteomics let Antares physically detect transient and allosteric binding sites, then design against them - grounding its machine learning in experimental reality.
Rather than chase crowded, well-drugged proteins, Antares prioritizes targets with strong human-genetics validation that remain unaddressed - aiming for first-in-class rather than fast-follower.
A near-$1.9B Novartis collaboration before a single clinical candidate is a market signal few platform peers can point to this early.
Where it fits: Antares sits among precision-oncology and undruggable-target players such as its former parent Scorpion Therapeutics, Kymera, Nurix, Relay Therapeutics, and Foghorn Therapeutics - competing on the strength of its integrated discovery engine and its focus on pockets that appear only when proteins move.
The precision-oncology company that would later spin out Antares is established.
The PI3K-alpha divestment sets the stage for Antares to house Scorpion's remaining investigational assets.
The company debuts as a spin-out focused on undruggable targets, backed by a top-tier investor syndicate.
A partnership worth up to ~$1.9B, with $105M upfront, targets undruggable cancers.
The most advanced program is slated to enter clinical testing.
It discovers and develops first-in-class small-molecule precision medicines against historically undruggable targets in cancer and other serious diseases, using an integrated platform of medicinal chemistry, target biology, and computational/predictive sciences.
It is a 2025 spin-out of Scorpion Therapeutics, formed to advance Scorpion's remaining pipeline after Scorpion sold its lead PI3K-alpha program to Eli Lilly. The name references Antares, the brightest star in the Scorpius constellation.
Antares launched in June 2025 with a $177 million Series A led by Omega Funds, Atlas Venture, Lightspeed Venture Partners, BVF Partners, and Cormorant Asset Management, among others.
The June 2026 collaboration includes $105 million upfront and up to roughly $1.9 billion in total option, development, regulatory, and commercial milestone payments, plus tiered royalties on net sales.
The company's most advanced program is expected to enter clinical testing in 2026.
Profile compiled from public sources including company announcements, BioSpace, Endpoints News, Fierce Biotech, PitchBook, and Crunchbase. Figures are approximate and reflect reporting as of mid-2026.