Lit university research lab window facing a financial district skyline, early-stage biotech funding gap | Healthcare Venture Capital Fund

Who Will Fund Discovery? Healthcare Venture Capital’s Stage Zero Problem

Healthcare venture capital has learned to buy innovation that already works. The harder question for 2026 is who pays for the experiments that produce the next generation of it.

Biotech’s money came back in 2026. It just did not come back to the beginning. Total venture dollars rose 22 percent in the first half of the year, BioCentury reported, while the number of seed and Series A rounds fell to its lowest first-half level in a decade and the capital behind them dropped 15 percent. BiopharmIQ, which tracks private funding events, counted about $3.97 billion of pre-seed, seed and Series A funding across 137 events in the first half of 2026, down from roughly $5.4 billion across 198 events a year earlier.

The headline rounds, meanwhile, keep getting larger. Isomorphic Labs, the Alphabet company founded by Demis Hassabis to apply DeepMind’s AI to drug discovery, raised $2.1 billion in a Series B in May. Capital is concentrating in companies that already have data, platforms or clinical programs, and thinning out at the stage where new biology gets found.

Call that stage stage zero: the work that happens before there is an asset to license, a molecule to dose or a milestone to price. Jeremy Levin, the former Teva chief executive and founder of Ovid Therapeutics, makes a related case in his 2026 book, Biotech in the Balance: Saving a Strategic Industry in an Age of Distrust: biotechnology is a strategic national asset, and investors need the nerve to back it for the long timelines the science demands. This piece looks at the same problem from the investor’s side of the table: where the capital is going, why each decision makes sense, and who is positioned to fund what the market now skips.

Capital Follows the Buyer

Venture capital is priced off the exit. A fund raises money for roughly a decade, invests it in companies and returns cash when those companies are acquired or go public. So venture investors watch closely what large drugmakers will pay for, and the answer right now is assets with human data. J.P. Morgan’s review of 2025 found that seed and Series A biotech capital fell to its lowest level this decade, about $8.7 billion across 191 deals, while Series B and later rounds drew roughly $16 billion. The bank attributed the early-stage weakness to heightened diligence standards and longer decision timelines.

None of this is irrational. A molecule with Phase 1 or Phase 2 data has a visible path, a visible remaining cost and a visible buyer. A team chasing an unvalidated mechanism has none of those, and its first decisive readout may be five years away. For a fund with a fixed life and limited partners expecting distributions, the second kind of bet gets harder to justify every year it runs.

The New Supply of Ready-Made Medicines

The buyer’s options have changed too. A growing share of the drug candidates that large pharmaceutical companies license now come from China. Evaluate Pharma data reported by Axios projected China-sourced candidates at almost 40 percent of all licensing deals in 2025, up from less than 3 percent five years earlier. China’s drug regulator counted 157 out-licensing deals worth a combined $135.7 billion in 2025, and Chinese state media, citing the same regulator, put the first-half 2026 total at about $110 billion, a record, according to Reuters. IQVIA, counting therapeutic in-licensing deals on its own basis, put the first-half 2026 value at $92 billion, already 88 percent of the 2025 total.

These assets are no longer bargains. Evaluate data shared with Fierce Biotech show the average upfront payment in a licensing deal between a Western drugmaker and a Chinese biotech rose about 230 percent, from $52 million in 2022 to $172 million in 2026. Buyers are paying for speed and evidence: molecules that have already cleared early testing in people.

For a US or European venture investor, that is direct competitive pressure. A preclinical company raising a Series A now competes for a pharma buyer’s attention with clinical-stage molecules from Shanghai and Suzhou. The rational response is to back fewer new discovery teams and more companies that already hold a clinical asset, whether developed at home or licensed from abroad.

Rational Deals, Irrational Totals

Each of those decisions holds up on its own. The problem only appears in the aggregate. When pharma buys proven assets, venture funds build toward proven assets. When venture funds build toward proven assets, founders design companies around the milestones a buyer will pay for, and fewer teams form to explore biology nobody has validated. No one decides to underfund discovery. It happens as the sum of sensible choices.

The cost shows up years later, because discovery is where future assets come from. A 2018 study in the Proceedings of the National Academy of Sciences traced the published research behind all 210 new molecular entities the FDA approved from 2010 to 2016. NIH-funded research was associated with every one of them, with project costs above $100 billion, and more than 90 percent of that funding concerned the biology of the drugs’ targets rather than the drugs themselves (Cleary et al.). The molecules buyers compete for today rest on target biology someone paid to understand a decade or two earlier. For a closer look at how research money is spent before it reaches the bench, see our analysis The Overhead Waterfall: Where a Research Dollar Actually Goes.

There is also value that never becomes an asset at all. Discovery organizations produce failed experiments that reveal new biology, scientists who carry hard-won skills into new companies, and tools that prove useful in fields nobody planned for. Those spillovers are real but almost impossible to price on a term sheet, which is exactly why a market organized around term sheets tends to underfund them. Our research partner Healthcare Discovery traces several of those stories, from a desert lizard’s venom to a Yellowstone hot spring, in its companion piece, The Discovery Gap.

Who Has Funded Stage Zero Before

When conventional venture capital has stepped back from early discovery, other kinds of capital have stepped in. The most instructive case is the Cystic Fibrosis Foundation. Beginning in the late 1990s, the foundation adopted what it called venture philanthropy: putting money directly into drug developers to absorb early risk that industry would not take on. Over two funding rounds it provided about $150 million to Vertex Pharmaceuticals for its cystic fibrosis program. That work produced Kalydeco, approved by the FDA in 2012 as the first drug to treat the underlying cause of the disease rather than its symptoms.

In 2014 the foundation sold its royalty rights on Vertex’s cystic fibrosis drugs to Royalty Pharma for $3.3 billion and said it would reinvest the proceeds in research and patient programs. Its chief executive at the time said Vertex itself credited the foundation’s early, de-risking money with making the drugs possible. The model drew criticism over the price of the resulting medicines. As a financing structure, though, it did something the venture market struggles to do: it paid for the risky front end with patient, mission-driven capital and still earned a return large enough to fund the next round of discovery.

A newer version of the idea is the Arc Institute in Palo Alto, launched in December 2021 with more than $650 million in pledges from founding donors including Patrick and John Collison, Vitalik Buterin, and Dustin Moskovitz and Cari Tuna. Arc funds its core investigators for renewable eight-year terms so they can pursue high-risk research without applying for outside grants. That is the opposite of a fixed-life fund chasing milestones: long, flexible money aimed at questions whose answers cannot be scheduled.

Where Family Offices Fit

Family offices sit closer to those two models than most venture funds do. Many have no fixed fund life, no fundraising cycle and no limited partners expecting distributions on a schedule. That makes them one of the few pools of private capital able to underwrite discovery on discovery’s timeline. It also lets them do things conventional funds rarely do: back a scientific founder before there is an asset, co-fund research alongside a disease foundation, or hold a small position in a discovery platform for ten years.

The discipline is in sizing and structure. Discovery bets should be small relative to the whole portfolio, spread across several teams, and paired with holdings that pay predictably while the science matures. That is the logic of the barbell many healthcare investors now use: long-dated, high-variance exposure to discovery on one end, and steady contractual cash flow, such as net-leased medical real estate, on the other. Our affiliate Healthcare Real Estate Fund covers the income end of that barbell.

Questions worth asking before backing an early-stage healthcare fund or company:

  • Where does the capital go? What share of the fund goes to first financings and company creation, versus later rounds in companies others started?
  • Is the thesis discovery or in-licensing? Licensing assets from elsewhere and generating new biology can both work, but they are different businesses with different risks.
  • How far does the money reach? How long can the company run before its first decisive experiment, and is the round sized to get there?
  • Who else is at the table? Disease foundations, academic partners and strategic investors alongside venture capital are often a sign of real unmet need.
  • What survives a failure? If the lead program fails, does the team keep its data, tools and learning? Organizations that keep discovering are worth more than single shots.

Watching the Front End of the Pipeline

The data that shows whether discovery is being funded is public but scattered: Form D filings, seed and Series A announcements, licensing terms and foundation grants. Healthcare Venture Capital Fund follows the transaction side in its healthcare M&A deals database, while Healthcare Discovery serves as the research and intelligence layer, following the science before it becomes an asset. Read together, they show where capital is flowing and where it is not.

The industry does not lack money. It lacks money willing to wait. For investors with the patience to fund stage zero, that scarcity is the opportunity.

Healthcare Discovery Tracks

Healthcare Discovery is the research and intelligence layer behind Healthcare Venture Capital Fund. It follows medicine from the lab bench to the exam room:

  • New discoveries: the Daily Rounds brief and in-depth coverage of the studies, drug candidates and AI-driven research moving medicine forward, starting with The Discovery Gap.
  • Doctors: a nationwide doctor directory built from more than 1.4 million public provider records, including Drug and Device Payments data. How the directory is built.
  • Healthcare companies: a company directory of venture-backed startups across biotech, medical devices and digital health.
  • Medicare innovation: the Medicare ACCESS Model tracker, listing every company approved to deliver technology-enabled chronic care.
  • Health claims: scored against the published evidence with the HYPE Index.

Explore Healthcare Discovery

Sources

This article is for informational purposes only and is not an offer to sell or a solicitation of an offer to buy any security or interest in any fund.

Similar Posts