For more than seven years the largest flotation in biotechnology by a venture-backed company was Moderna's, which raised $604m in 2018. The record stood through a pandemic, a boom and a three-year drought. It was broken twice in two months.
In April 2026 Kailera Therapeutics raised $625m. In June, Parabilis Medicines priced 33.5m shares at $20 and took $670m, closing its first day at $31.60, some 58% above the offer. By midsummer 13 venture-backed biotechs had gone public in the United States, raising $4.5bn between them at a median of almost $302m each, and more than half had banked at least $300m. Most were trading above their issue price.
This was supposed to settle an argument.
Since 2022 the standard account of the reverse merger in biotechnology has been that it is what companies do when the IPO window is shut. Failed listed biotechs accumulate, private companies cannot go public conventionally, and the two find each other. On that account the wave depends on the window staying closed. Open it and the structure should drain away.
The window opened at record scale and the structure stayed. On June 8th, the day before Parabilis priced, Treeline Biosciences announced a merger into Standard BioTools that will bring it public with more than $900m in cash. The two records were set a day apart.
Two products, not one
The prediction failed because the substitution it assumed was never real.
An IPO is a marketed offering. Underwriters build a book, price is discovered among public investors, and the company submits to a fortnight of weather it cannot control. The alternative is a negotiated private round whose last step happens to be a ticker. The buyers differ, the pricing mechanism differs, and the moment in a company's life that suits one rarely suits the other. Parabilis, with a lead asset the market could see and value, was never going to take a shell. Treeline, which spent years in deliberate obscurity, was never going to run a roadshow.
Both routes can expand at once, and in 2026 they did. Watching the biotech IPO market for a signal about the other channel is watching the wrong instrument.
What was supposed to deplete
The other half of the cyclical case was that the specialist bid would thin, since crossover funds were said to prefer negotiated entry only because public offerings had treated them badly.
Stifel's Biotech Buyside Study, built from Form ADV, 13F, 13G and 13D filings and current to May 2026, counts 237 specialist life-sciences funds holding $464bn of biotech assets. It counts separately 158 healthcare-only specialists, managing $353bn, against 94 such funds in 2024. Across all tracked fund types, biotech holdings reach $1.1trn, of which index and quant money accounts for $423.3bn.
The private-placement market beneath that has industrialized. Over a 40-month window, 50 funds took part in nine or more placements or registered directs of $50m or more. RA Capital did 53. Janus and Adage did 43 apiece. That is the cadence of a standing desk with a process, running the same transaction several times a quarter.
What did change
Something did shift over the period, though not in the direction the cyclical story predicted. The structure became slower, and more like the thing it was supposed to be an escape from.
Nasdaq's Rule 5110(a) requires a listed company that changes control to re-qualify for initial listing, and the review has grown teeth: the full quantitative standards, and qualitative ones covering governance, internal controls and operational continuity. Procopio put the shift in a sentence in November 2025. These deals today, it argued, "are not about speed. They are about readiness."
The Securities and Exchange Commission has pushed the same way. Its Rule 145a, adopted in January 2024 and effective from July that year, deems a combination between a reporting shell company and an operating business to be a sale of securities to the shell company's own shareholders, and staff have read shell-company status broadly enough to capture failed biotechs whose real contribution is cash and a listing. A company caught that way loses the use of Form S-3 for a year. Treeline's deal carries a contingent value right for Standard BioTools' legacy holders tied to the sale of its mass cytometry and microfluidics businesses, which is exactly the feature the staff treats as evidence.
So the advantage that once justified the structure, that it was quick and lightly examined, has largely gone. The deals kept happening anyway. That is the strongest single argument that something structural is at work.
The tell is the engineering
Consider what dealmakers have built to keep the transaction working.
Nasdaq Rule 5635 requires a shareholder vote before a listed company issues common stock amounting to 20% or more of the shares outstanding in an acquisition, and before any issuance that hands over control. In these deals the incoming shareholders take far more than that. Rather than wait for a proxy, counsel now structure the excess as non-voting convertible preferred stock, which converts to common once shareholders approve, letting the parties sign, announce and close on a single day. Orrick's Tony Chan and David Schulman set the mechanism out in 2024: the rump above the 19.9% line, in Schulman's phrase, "ends up being in a non-voting convertible preferred".
Markets do not engineer around a rule for a transaction they expect to stop doing. The sign-and-close structure is infrastructure, and infrastructure implies an expectation of volume.
The exception in a concentrating market
A pattern runs underneath all of this, and it has nothing to do with which door a company uses.
The 2026 biotech IPO market is at once small and enormous: 13 companies, more than half of them raising at least $300m, at a median near $302m. Venture funding matches it. Biotech companies raised $9.1bn across 68 businesses in the first half of 2026, the strongest half since early 2022, and the strength came from a few very large rounds, not from more companies getting funded. At the other end of the market, seed and series A activity contracted to roughly 50 deals worth $2.3bn in the first quarter, against 60 worth $3.7bn a year earlier. Why that end is being squeezed is contested. Bankers point to more diligence, slower decisions and a preference for assets that clinical data has already de-risked.
Capital is concentrating into fewer and larger commitments. One route did not thin with the rest. William Blair counted nine of these mergers in the second quarter of 2026 alone, carrying roughly $1.6bn of concurrent private placements, after five in the five months to March at a median raise of $200m. More deals, and bigger ones, in the quarter the flotation market set records.
It also shows who is left out. The price of entry has climbed. Yarrow Bioscience's merger in late 2025 came with $200m, Damora Therapeutics with a $285m placement in the same quarter. Treeline will carry more than $900m of pro-forma cash onto the exchange, and needed no concurrent placement at all. The question for a mid-sized private biotech is no longer whether the window is open. It is whether the company is large enough to be admitted through any door at all.
Where the cyclical case still stands
The other side deserves a hearing. After this year's data it is thinner than it was.
Its best evidence is old and indirect. In 2015, in a low-rate world with an open new-issue market, only 79 companies across every industry went public by this route, the fewest since tracking began in 2008. The route did fade last time conditions were kind. Nobody knows whether biotech did the same in 2020 and 2021, because no one has published the count.
Nor has anyone re-benchmarked the supply. The most recent usable estimate of biotechs trading below cash comes from a single analyst's blog post in November 2023, which put it above 30% of listed biotechs without naming a data source or defining the universe. Whether that pool has shrunk or refilled since is unknown, and it matters more than any other figure in this argument. No quarterly series exists.
What the cyclical case can no longer claim is the deal count. DealForma counted 31 of these transactions and SPAC listings in 2023, 13 of them SPACs, and 18 in the first three quarters of 2024. Set against William Blair's nine in the second quarter of 2026 alone, the route is thickening as the window opens.
What to watch instead
If the IPO window is the wrong instrument, there is a better one, and it is unglamorous.
The whole structure rests on a single clause. Nasdaq defines a reverse merger as an operating company combining with a shell, then excludes from that definition the change-of-control combinations described in Rule 5110(a). That exclusion is why a private biotech taking over a listed one avoids a year of seasoning and reaches the market at once. Remove it and the arithmetic of every deal described here changes.
Nobody has proposed removing it. Nasdaq has been busy in this corner of the rulebook: in December 2025 it narrowed the same definition to let over-the-counter blank-check vehicles out of the seasoning requirement when they list on an effective registration statement, and in May 2026 it imposed a new $25m public-float test on Rule 5110(a) combinations, though only for China-based issuers. The drafting is under active attention. The biotech route was left alone. Searched thoroughly, that absence is the finding: no filing at Nasdaq, the New York Stock Exchange or the commission proposes removing the carve-out.
Which answers the question this piece began with, though not in the form it was asked. The wave does not end when the IPO window opens, because it never depended on the window being shut. It ends when someone amends a definition. The thing to watch is not a market. It is a paragraph.