Biotech Venture Capital: Why Later-Stage Draws Dollars and Early Formation Is Lagging
Biotech venture capital has gone from euphoria, to hangover, to something closer to sobriety. The sector isn't broken; it's back to fundamentals and less tolerant of risk. Drug discovery and preclinical development are bearing the brunt as capital chases clinical proof and the utilization of Artificial Intelligence (AI) in the biopharmaceutical industry.
Between 2019 and 2021, record-low interest rates and the urgency of COVID-19 therapeutics and vaccines drove an unprecedented boom in biotech funding, M&A and IPOs. Global and US VC funding in life sciences peaked around 2021, with US life sciences VC hitting roughly $47B that year before starting a multiyear slide1. Public-market indices followed the same arc: the XBI biotech ETF fell about 61% from its February 2021 high through 2023 before finally bottoming and recovering in late 20232.
By mid-2025, the sector looks very different – very slow but not dead. It's a smaller, pickier, more data-driven market that creates real pain for many companies and real opportunity for investors who actually understand the science and the teams that can deliver.
From Boom to "Biotech Winter"
After the 2021 peak, the funding cycle turned decisively. Venture investment and IPOs fell sharply; by mid-2025 "almost no new biotech offerings" were reaching the public markets, and layoffs and pipeline cuts have rolled through the industry since 2023 as companies conserve cash3.
On the private side, total US VC dollars into life sciences dropped from 2021's highs to roughly $25B in 2023 before a modest uptick to ~$27B in 2024 – still well below the boom-era run rate1. Capital raised by dedicated biotech VC funds also shrank, from about $30.8B in 2021 to $11.7B in 2024, as higher interest rates made it harder for GPs to raise new vehicles4.
Yet, at the same time, the size of individual successful financings has remained large. Q1 2025 data show the median US biotech venture round hovering near $100M, with a persistent trend toward "megarounds" for a select set of companies5. In 2024 there were nearly 100 nine-figure financings, heavily skewed toward companies advancing Phase II programs6.
The picture that emerges is not of a frozen market, but of a barbell: ample capital for a relatively small number of perceived winners, and a long tail of companies struggling to raise anything at all.
Where the Money Went
What changed after the 2021 peak wasn't just risk appetite; it was competition for home-run returns. For a decade, biotech and biopharma were the default destination for deep-tech VC chasing asymmetric upside. But in 2023–2025, Artificial Intelligence (AI) emerged as the new gravitational center for venture dollars.
Global funding to AI startups hit a record ~$100.4B in 2024 – with an all-time-high $43.8B in Q4 alone – driven by a handful of massive rounds into model labs and AI infrastructure. In 2025 year-to-date, PitchBook/NVCA show AI capturing over half of VC dollars (and an even larger share in the U.S.), reflecting a structural reallocation of risk capital toward AI's perceived speed to scale and Total Addressable Market (TAM).
LPs allocating to venture have finite risk budgets, and AI's speed to massive valuations is reshaping how they think about biotech's 10–15 year return cycles. When a frontier AI company can go from Series A to $10B+ in 24 months, the patience required to get a biotech from target discovery through Phase III looks like a bad bet to generalist investors. The capital that remains in biotech is more specialized: funds and LPs who understand the asset class and will underwrite long timelines. The tourists have left.
Meanwhile, U.S. life-sciences VC has reset from the 2021 boom ($49.3B) to ~$26.3B YTD through Q3 2024, i.e., back near 2018–2019 levels. The result is a market where biotech quality still gets funded, but the easy beta bid is gone because AI now soaks up the incremental risk budget19–22.
The 2020–2021 surge didn't just raise prices – it pulled forward a larger late-stage cohort. Citeline's 2024 review shows record pipeline size with more drugs in Phase I/II year-over-year, and the pandemic-era IPO wave left more public biotechs working through mid- and late-stage trials than a "no-boom" world would have produced. With capital recalibrated, it's rational that more dollars chase later stages – there's simply more inventory there23–25.
VCs Have Pivoted Hard Toward Clinical Data
In this environment, investors have become much more selective and are explicitly migrating toward "de-risked" assets. Q2 2025 commentary from multiple law firms and banks notes that VCs are "favoring de-risked, later-stage assets" while earlier companies endure an extended "biotech winter"7.
Analyses of recent financings show capital concentrating in Phase II and other clinically validated programs, with median venture round sizes for Phase II companies rising from ~$50M in 2024 to ~$63M through Q3 20258,9. Clinical data – especially mid-stage data with a clear regulatory path and visible M&A comps – has become the primary underwriting lens. New funds are being raised that explicitly focus on late-stage clinical trial financing and royalty-based structures, such as Abingworth's Carlyle-backed $1.5B clinical trials fund10.
In contrast, new company formation is happening, but more cautiously. Early-stage deals (Seed/Series A) have not returned to the frenetic pre-COVID pace. 2024–2025 data suggest early-stage volumes remain below pre-pandemic levels even as deal sizes for the best stories have grown, and there are signs that even those Series A sizes are flattening11.
Looming NIH budget pressures – potential cuts of up to $20B – are more than a headwind for university spinouts and early platform companies. If those cuts happen, it's reasonable to expect the academic pipeline that feeds novel company formation will contract for a considerable period. This isn't cyclical; it's a potential structural break in how early biology gets funded. The effects will likely lag the cuts, creating a false sense of stability before the pipeline actually thins12.
Taken together, VCs are behaving less like "company builders" and more like late-stage stock-pickers: waiting for proof and piling in once the curve has bent.
The Quiet Risk: Late-Stage Companies on Shaky Science
One uncomfortable reality of the boom years is that a non-trivial number of later-stage biotechs were built on scientific hypotheses that were never truly derisked – protected for a time by cheap money and a forgiving market. However, we are now seeing the consequences: 2024–2025 began with "an unsettling series of high-profile clinical failures" even as megarounds continued for a different cohort of Phase II companies6.
Additionally, industry norms for late-stage success are modest – 55–60% success rate for the industry at large – implying that 40%+ of late-stage programs still fail10. That baseline attrition rate assumes programs advanced on scientific merit. But a meaningful subset of today's late-stage biotechs seem to have gotten there because capital was available, not because the biology was compelling. A higher-than-usual rate of costly failures over the next several years wouldn't be surprising.
For investors, that's both a risk and an opportunity: many public and late-stage private companies are mispriced because the market either underestimates or overestimates the true probability that the underlying biology will work.
Government Scrutiny on Pricing and Regulation Is Real
At the same time, long-term return assumptions are being reshaped by policy.
The US Inflation Reduction Act (IRA) of 2022 created, for the first time, a mechanism for Medicare to negotiate prices on selected high-spend drugs and to claw back rebates when prices rise faster than inflation13. The first set of negotiated prices will take effect in 2026 and is expected to shave billions from both beneficiary out-of-pocket costs and overall Medicare spending14,15.
Drug makers have challenged the program in court, but early appellate decisions have largely upheld the government's authority to negotiate, rejecting arguments that companies have a "right" to sell to Medicare at pre-negotiation prices16.
For VCs and their portfolio companies, this doesn't mean innovation is no longer rewarded – but it does mean that:
- Out-year pricing power is less certain, especially for chronic, high-spend drugs that become IRA targets.
- Traditional "lifetime peak sales" models need to factor in earlier and steeper price erosion. This will be particularly important for second- and third-in-class molecules, unless they are truly differentiated from the leader.
- Marginal late-stage programs with weak differentiation are more exposed, because there is simply less room for error on price.
- The therapeutic areas most exposed are those with large Medicare patient populations and established treatment paradigms where new entrants offer incremental rather than transformative benefit: chronic cardiometabolic conditions, oncology supportive care, and certain CNS indications where payers already have generic alternatives.
Regulatory and pricing scrutiny is very real, and it will disproportionately impact commodity-like assets that don't deliver step-change clinical value.
Where The Opportunity Is Now
If you strip away the noise, today's market actually looks attractive for disciplined, science-led investors.
1. Valuations have reset, but biology hasn't changed. Biotech funding is down sharply from 2021, public valuations remain far below the peak, and some listed investment vehicles are even winding down2,3,17. For investors with fresh capital, this is the classic "buy quality in a downturn" setup.
2. Capital is concentrated, not absent. Large megarounds for strong Phase II assets show that when the science and data are compelling, capital appears in size5,6,8,9. That's a powerful signal: the market is not closed; it's just no longer funding everything.
3. There is a gap in true early-stage company creation. Many traditional VCs have pulled back from ground-up company formation, especially for novel biology that lacks a near-term clinical readout. At the same time, platform innovation (e.g., AI-enabled target discovery, new delivery modalities) continues to attract attention when it's connected to clear clinical strategies12,18. This opens space for specialized groups that can combine deep translational science with disciplined, milestone-based financing.
4. Distressed and stranded assets are everywhere. The wave of restructurings since 2023 has left many programs "orphaned" inside public micro-caps or under-capitalized privates3. Careful scientific re-underwriting can identify which of these assets are fundamentally flawed – and which are victims of capital cycles and can be acquired, recapitalized, or repurposed.
5. Policy risk rewards true differentiation. In a world of negotiated pricing and tougher scrutiny, assets that deliver large, hard-endpoint clinical benefits – or that address clearly underserved populations – are more likely to command durable value. Conversely, late-stage programs built on marginal differentiation and flimsy biology will be squeezed from both sides: clinical risk and pricing pressure.
Where This Leaves Us
The mood in biotech feels gloomy. Fundraising is slower, IPOs are rare, and the headlines favor layoffs over launches. The current state of venture capital in biotech/biopharma is not the free-wheeling optimism of 2020–2021, nor the outright despair some painted in 2022–2023. Zoom out and this looks less like a broken market and perhaps more like historic norms with some distinct caveats – at least by pre-2020 standards. The pandemic era created a distortion: money was cheap, and urgency was high. Now, capital is choosier and more expensive and the bar is partially back where it belongs – on data.
That shift stings most for preclinical companies, which are paying for the whiplash of the last cycle as capital has shifted to later-stage, de-risked clinical programs. Even after the reset, late-stage attrition remains a fact of drug development. The boom likely advanced some programs faster than their biology warranted, and we may expect a higher-than-usual rate of disappointing readouts across select public names over the next few years. A simple reminder that biology and trial design, not momentum, determine outcomes.
Compounding this all is Artificial Intelligence (AI). For years, if you wanted true venture-scale upside in deep tech, you went to biotech. Today the AI boom is absorbing a massive share of risk capital and attention, and it's reshaping how LPs think about biotech. The net effect isn't that biotech is unfundable; it's that the easy bid is gone and the generalists have moved on.
The sector is sober again, which is when the returns actually get made. For investors who can actually interrogate mechanisms, pick winning teams, understand translational risk, and navigate a tougher policy backdrop, this is precisely where excess returns are born.
The Bottom Line
If you strip away the noise, the opportunity is hiding in plain sight. The market is systematically underweight in early company creation – not because the science isn't there, but because the crowd has migrated to mid-stage "de-risked" assets or to AI altogether. That creates a window for milestone-driven, stage-gated investing at formation and through early clinical proof.
The winning strategy over the next few years may look boring on paper, but powerful in practice: underwrite fewer programs, insist on crisp translational packages, and tie financing to objective clinical readouts. The firm that quietly accumulates a portfolio of Phase 1/Phase 2-ready assets while others chase later-stage momentum will have real optionality – sell, partner, or scale – when the cycle turns.
About the Authors
Lucas Rodriguez, PhD is a Partner at Alacrita focused on investor support and fundraising strategy. A former biotech founder and CEO, Lucas co-founded CerSci Therapeutics and led it through IND-enabling studies, Phase I trials, and successful acquisition by Acadia Pharmaceuticals (NASDAQ: ACAD). He previously served as CEO of Doloromics, a precision medicine company developing novel therapeutics for chronic pain. Lucas holds a PhD in Biomedical Engineering from The University of Texas at Dallas.
Mark Philip, PhD, MBA is a Partner at Alacrita with over 30 years' experience in the biopharma, medical device, and diagnostics industries. A serial CEO and board member across multiple startups and large businesses, Mark has orchestrated turnarounds, built product pipelines, and executed company sales across the US, Europe, and Asia. He is on the faculty at Boston University's MBA Program, teaching advanced strategy in pharma and medical devices. Mark holds a PhD in Stem Cell Research from Nottingham Trent University and an MBA from Lake Forest Graduate School of Management.
Simon Turner, PhD, FRSC is a former Managing Partner and current Strategic Advisor at Alacrita with nearly three decades of experience in life sciences strategy and product development. A former CEO of Biotica Technology, a venture-backed drug discovery company focused on microbial synthetic biology, Simon has worked as an entrepreneur, executive, venture capital investor, and advisor. He previously spent ten years as a management consultant at Arthur D Little and Andersen Consulting. Simon holds a PhD in Organic Chemistry from the University of Cambridge and is a Fellow of the Royal Society of Chemistry.