Over 200 GLP1 obesity programs are in development today.  More than 300 PD1/PDL1 cancer programs exist.  Greater than 150 programs target each of CD19 and KRAS. Therapeutic crowding has ratcheted up the competitive intensity in biotech to new levels.

While a perennial concern for decades in the biopharma industry around “hot” mechanisms of action (e.g., statins and SSRIs in the 1990s, TNFs in the 2000s, PD1’s in the 2010s, GLPs today), this herding into well-established mechanisms has accelerated in recent years – both in the volume of activity and in the pace of competition.

Here on this blog we’ve covered these topics a number of times in the past, with the “March of the Lemmings” in cancer in 2012 and “I/O: The Strategic Supernova In Cancer Today” in 2016.

But with the rise of China’s innovative sector, the proliferation of biologic drugs coded by DNA, and more than dozen years of abundant capital, to name just a few things – the challenge has only gotten worse.

As illustrated in LEK’s May 2025 report “Is Biopharma Doing Enough to Advance Novel Targets?” the scale of the current crowding challenge is staggering. All of these targets have at least 50 programs against them in preclinical or clinical development.

Further, the pace of “fast follower” drugs is striking relative the past: according to PhRMA’s recent report on innovation, the average time for a drug class to have three FDA approvals in it went from ~15 years in 1990-2003 to ~2 years in 2013-2021.

Many others have discussed this “therapeutic crowding” topic, so I won’t belabor the points here.  They have already explored and indicted it for a myriad of negative impacts on the sector: poor biotech success rates, failures of IPOs and Smid-cap biotechs, lack of generalist investors in the space, rise of short sellers, and the demise of the XBI biotech index, amongst other bad karma for the sector.

But who is to blame?  VC’s, obviously.

Indeed, VCs are commonly cited as the culprit: we apparently start and fund too many companies doing the same things.  Too much capital chasing too few novel ideas.  Too many “me-too” programs.

Since 2012 when the JOBS Act primed the IPO pump, we’ve brought over 600 companies public on US exchanges, many of which doing similar things to other biotechs. Think about how many CART CD-19 companies got started… or gene editing and gene therapy plays… literally, dozens and dozens of them emerged, all chasing similar indications. During 2020-2021, the bubble in venture creation started over 180 biotech’s a quarter in the US alone (which thankfully is down to a third of that recently). Private biotech investors, including and frequently led by VCs, were behind most of these companies.

In light of all that, the criticism undoubtedly has some truth to it. VCs are in part responsible for startup biotech’s contribution to the rise in therapeutic crowding (though Pharma and other established players also do their part to contribute to this dynamic as well).

But we can’t throw the baby out with the bath water: yes, VCs helped crowd the space – but they also bankrolled the breakthroughs. No crowd, no drugs. Without early stage investors, biotechs couldn’t really get started, as the sector depends on them for capital at a minimum, and strategic value-add and corporate governance on a good day.

So here’s my contrarian counterpoint: if you believe in markets and capitalism, you have to embrace the competitive tension that crowding manifests. May the best companies with the most successful drugs win. Fail to deliver valuable and differentiated medicines and your shareholders will face losses.

In the long run, this dynamic marketplace of rewards and losses creates a mechanism for bringing great drugs forward – and I’d argue this system is good for patients around the world. There are some real downsides, which I’ll mention below, but it is risk-taking capitalism that delivers innovation.

Crowding around new innovations also isn’t a disease only biotech VCs can succumb to: all VCs suffer from a powerful herding instinct.  In fact, (over-)allocating capital to “hot” ideas is wired into the DNA of the venture capital industry and the market cycle of innovation (euphoric peaks, despairing troughs).

The tech VC sector, nearly 10x the scale of biotech, does this all the time.

Take ride-hailing app’s. Uber and Lyft are the big global winners, but at least scores of startups were launched in the US alone, and upwards of 150 companies around the world.  Anyone remember Sidecar?  Started in 2011, before Lyft, ended up shutting down in 2015.  Or Juno?  It started 7 years after Uber.

Or grocery delivery services. At least 50-70 startups formed to execute on this business model. Instacart is clearly the dominant player, but it started 13 years after FreshDirect and more than 20 years after Peapod. Tons of others failed along the way. Add in the meal delivery firms and there have been a few hundred startups focused on getting food to people.

The list goes on. Dating apps. Social media startups. Online betting platforms. Electric vehicles. So many “hot” spaces, all of which got crowded quickly, backed by VCs who hoped to pick a winner.

If you were smart or lucky enough to pick one of the dominant winners, you were a VC hero. But the reality is most investors in these spaces likely lost money… like lemmings running off crowded cliffs.

Even with these investment losses, there’s one clear beneficiary of the crowding into these spaces: consumers. And the crowds often consolidate into a few winners that have become ubiquitous in the lives of most Americans.

I’d like to believe that in the long run, therapeutic crowding will deliver something similar. While investors in many deals will lose money, those that deliver value for patients will accrue returns over time.  And in each disease or indication, we’ll have a handful of approved medical options to give patients choices around what level of benefit-risk they would like, and to give payors choice around what level of cost they can support.

It’s instructive, though, to dig into some of the profound differences between tech and biotech investing to understand how crowding differentially impacts our respective sectors.

In many areas of tech investing, there are minimal barriers to entry: a good coder can fire out a product offering quickly. Time to market is fast, and getting critical consumer feedback allows a startup to iterate on its product quickly. R&D cycles times are days, weeks, and months. Market competition between startups happens very quickly, and often intensely. Well-known financial metrics assist in tracking the momentum of these plays (e.g., MRR, CAC, users, etc). A deep and broad pool of seed, early stage, and growth investors exist, allocating capital to the winners (and, over time, away from the losers). If a startup can achieve escape velocity, leading to a massive reduction in its cost of capital, it can scale even faster… and scaling creates dominant market positions, leading to consolidation and a “winner take all” dynamic. So crowding in tech dissipates into a few big winners over the course of time, often within five years.

This isn’t really how biotech works.

Barriers to entry have historically been very high, as deep scientific expertise and integrated capabilities are required (though this is changing as drug R&D expertise democratizes and commoditizes around the globe). R&D is intensely regulated to prevent harm (and prove efficacy), leading to prolonged cycle times measured in years, and often decades from idea to approved drugs in new technologies.

Direct competition between drug candidates doesn’t happen early in development; instead, competition really only occurs via capital allocation decisions of investors making implied comparisons around target product profiles (TPPs). True head-to-head competition between products is delayed often until approval. Scientific data is the only real currency of competition (clinical or preclinical), not any robust financial metric.

Lastly, scale advantages don’t really accrue until a decade into the company life cycle: either in very late clinical studies (global Phase 3s), and often not until commercialization. Scale may in fact work against nimble innovative research and early development. Further, premature scaling for a startup is quite counterproductive (huge burns for early stage companies), rather than helpful escape velocity.

All of this conspires in biotech to lead to an ecosystem where firms must persist for far longer in order to reveal their actual product profiles. Successful firms are able to credibly communicate the TPP of their programs, and figure out how to raise capital (take equity dilution) and do partnership deals (take asset dilution) in order to maintain adequate balance sheets to transform their candidates into actual drugs. If you can do it faster or cheaper (like what China has been doing recently, or AI/ML promises to do), that obviously helps create R&D advantages during that long innovation process.

Further, unlike tech, there’s a relatively small universe of investors allocating scarce funds to help these new and emerging firms navigate that R&D journey to market.  There’s also an absence of late stage, “less valuation sensitive” private growth capital, which essentially forces loss-making development-stage biotechs into the public markets. In tougher times, like today, more startups will starve during this journey than might otherwise; but in frothier times, more will survive than probably should. Because capital allocators are human, their fear and greed drive this market cycle.  And those emotions move due to comparative data packages (and TPPs) of R&D stage medicines, not financial metrics, which frequently leads to mispricing/dislocations in the market – all part of what makes biotech investing both an opportunity and a challenge.

This long duration crowding dynamic further means that “me-too” or less differentiated products will survive longer than they might in the tech sector – at times surviving all the way to drug approval. This occurs in large part because one only really knows for sure if a drug is a meh “me-too” product until mid- or late-stage clinical trials have read out, usually 5-10 years after a startup forms.

In a free market that is actively allocating capital to new startups, the creation and persistence of players in crowded areas is bound to happen – in part because no firm plans to make a undifferentiated product. I’ve never heard a pitch from an entrepreneur that declares “we’d like to make a second-in-class product”.  Everyone has a dream for why they think they can make either a first-in-class or best-in-class product. It’s always a new angle on affinity, potency, delivery, pharmacology, selectivity, dosing, etc… And entrepreneurs good at selling that dream will find investors to back them. But, in the end, most dreams don’t come true – though it often takes a long time in biotech to realize that.

Fortunately, to at least a modest extent, me-too or incrementally different products do actually have a place in the biopharma marketplace: they offer more choice for patients around convenience, tolerability, or efficacy, as well as potential competition for formulary or payor status, leading to pressure on net drug prices.

In summary, crowding is the expected consequence of an active market for capital allocation (and management teams’ time allocation), and happens in tech and biotech venture investing. While there are profound differences in how this herding instinct impacts the two ecosystems, it’s just part of the innovation cycle – and part of the creative destruction so important in risk capital markets.

There are, however, real downsides to over-crowding for biotech that are important to not gloss over, especially in specific disease areas and modalities: real opportunity costs of not funding out of favor areas with high unmet needs; patient participation and perhaps over-enrollment in trials for drugs unlikely to demonstrate value; dilution of scarce management talent over too many similar startups; and, long term investor losses leading to a shift (outflows) to other sectors, to name a few. These are real, and seeking to mitigate these risks is important, but in a free market for capital allocation they are unlikely to go away.

Startups crowding into hot spaces is axiomatic with markets. In a sector defined by science, R&D, and risk capital, crowding isn’t necessarily a bug – it’s a feature.  Winners will emerge from the crowd and bring new and differentiated drugs forward – and patients will benefit.

The equity markets have collapsed in 2025, the IPO window is closed, the FDA is in turmoil, the NIH is being gutted… and it’s a great time to start new biotech companies.

Why?  Because there are so few being created today and there’s far less competition for biotech’s startup resources.

Here’s a snapshot based on the current cut of Pitchbook data: the 1Q 2025 was the lowest quarterly level of new biotech startup formation in the US for a least a decade.

This trendline downwards has been in place for four years, since the all-time-high in 1Q 2021 when the $XBI hit it’s peak, and this contraction was the subject of prior blogs in 2023 (Biotech Funding: Times Are Tough, Maybe For The Better) and in 2024 (The Biotech Startup Contraction Continues… And That’s A Good Thing), both of which highlighted the value of shrinking the pool of VC-backed biotechs. The pace of new startup formation is now down nearly 70%.

If the ecosystem is facing such headwinds, and public investors in particular are fleeing to safer havens, why do we hold the contrarian view that now a great time to start new biotechs? Because we’ve been here before: over the past twenty years as a venture investor, I’ve witnessed multiple investment cycles.

Taking a purely microeconomic viewpoint, startups operate in an “biotech equity” supply-demand environment. When demand to purchase new equity accelerates in frothier markets, tons of supply gets created: over 170 startups got their first financings in 1Q 2021. When demand contracts, supply shrinks: as it has over the past four years. But as a supplier – which venture creation firms like Atlas are – we’d much rather operate in a world of scarce supply as we create our “products”.  When everyone is creating startups, hyper-competition for resources, patients, and mindshare is a challenge. But with scarce startup supply, when investor demand returns, which it will (as financial cycles are endemic to markets), we’ll see value appreciation: by providing a fresh supply of new equity ownership around promising drugs (with clean cap tables), founding/existing equity holders will be rewarded. Paraphrasing an early mentor of mine: as an early stage VC, you need to have an inventory of emerging investments for when the demand part of the cycle accelerates.

Beyond basic supply-demand economics, the three ingredients – or “resources” – required for VC-backed biotech startups are all very favorable today: science, talent, and capital.

Scientific substrate for startups is as strong and mature as ever.  As a sector, over the past decade we experimented with and developed a broad toolkit of modalities to address specific drivers of disease with a wide range of therapeutic deliveries. We are now deploying the tools that work best in the context of new medicines addressing real unmet needs, by developing degraders, drugs based on covalency and allostery, genetic medicines, or multi-specific engineered biologics, to name a few. And we’re sourcing this startup substrate from all over the globe (from China to the Cambridges, from Italy to Indianapolis) to create NewCo’s, often headquartered in our backyard biotech communities.

The market for talent has loosened considerably; great leadership and strong managers are never easy to attract and retain, but things are far more favorable for recruiting teams than back in 2021. Voluntary turnover rates are at decadal lows, and the pool of available talent is deep (and filling with the unfortunate RIFs and belt-tightening). While there’s always a “war for talent” for the best teams, the “heat of combat” has come down considerably in this market.

Private capital remains abundant by historic measures, even if more risk-averse today, and more focused on assets than platforms. The first quarter of 2025 saw more than $5B in venture funding for biotech, and over the past twelve months there’s been $25B invested. Compared to the same annual period in 2017 and 2018, this is 50% and 25% higher amounts of venture funding, respectively.  And it’s 300-400% higher than at the start of the secular bull cycles in 2013-2014.  Every week there’s a new $100M+ mega-round being announced. So there’s plenty of dry powder, even though it often feels like firms are just sitting on it. While private valuations for emerging companies seeking Series B and later private rounds remain challenging, funding rounds are getting done (though taking longer to close). And those backlogged private companies waiting to go public need to weigh the deep discounts of raising in the private markets (and the significant equity dilution that implies) with the alternative of selling/partnering assets to strengthen their balance sheets. We’ve seen a number of those deals lately.

Stepping back and looking at these long-term dynamics when you are in the midst of the bear market part of the financial cycle is often very difficult. And for investors (and management teams) who are being judged on their monthly or quarterly returns, it’s brutal. These near-term challenges are not to be taken lightly as they have ripple effects and consequences. Redemptions to funds and capitulation to stock prices raises the cost of capital, sometimes beyond existential levels – as we’re seeing in the public markets today.

Big doses of discipline are important medicine for emerging companies to take right now: focusing on portfolio priorities, tightening budgets and belts, exploring partnership alternatives, considering creative mergers, etc….

Multi-year contractions like the one we’re in help to recalibrate the overall number of biotech companies. Flux in this ecosystem determines the equilibrium state: startup creation adds to the system, exits and failures subtract.  Startup creation is down considerably, and while good exits are constrained (IPOs and mergers/acquisitions), failures have accelerated (shutdowns and liquidations/bankruptcies). So the “flux” favors a smaller overall ecosystem, though this process takes years to reset. Likes pigs going through the snake, biotechs work their way through the system. After four years of this prevailing flux dynamic, the private VC-backed ecosystem that remains is much healthier: the average health of the herd goes up with scarcity.

But it’s also important to remember that great companies are often born out of tougher times. Alnylam was started in the nuclear winter of 2002. Nimbus was formed in the spring of 2009, at the bottom of the markets. Kymera was being formulated in the late 2015/early 2016 bear market.

The startups that are being created today – the few and the proud – will likely be the emerging stars of 2030’s. It takes years to go from idea to clinical proof of concept in patients; unless you’re a startup being formed with existing drug candidates, it’s unlikely a de novo discovery startup will have true clinical PoC before the end of the decade (only five years away). As an early-stage venture investor, a core part of our job to help enable these future winners – by creating, funding, building and scaling the next generation.