Biotech & Longevity · Wednesday, 24 June 2026
01 · Briefing · what happened
The week three big drug bets failed — and what the winners' headlines hide
Pfizer, Exelixis and Novocure all saw late-stage trials fall short this week, even as the LSD pill and the $10.9bn deal grabbed the cheers. A normal week in biotech is mostly failure — we just rarely hear about it.
Key takeaways
- Three big late-stage drug bets failed this week — Pfizer's lung-cancer drug, Exelixis' colorectal drug, and Novocure's brain-cancer device — while the LSD depression pill and a $10.9bn deal took the headlines.
- Of drugs that reach human testing, only about one in ten is ever approved; in cancer it's worse — so a week with this many failures is the field's normal, not a crisis.
- We mostly hear about the drugs that work, which makes medicine look like a steady march of wins and hides how often the bets quietly fail.
It was a good week to be a biotech headline and a bad week to be a biotech drug. The cheers went to a depression pill made from LSD, a $10.9 billion takeover, and a gene therapy heading for approval. Out of the spotlight, three closely-watched, expensive bets quietly fell apart.
The drugs that failed
On Monday, Pfizer said its experimental lung-cancer drug, sigvotatug vedotin, fell short in a phase 3 trial — the big, final test before approval, where a drug is compared against the current standard in hundreds of patients
The same week, Exelixis said its drug zanzalintinib missed its main goal in a colorectal-cancer trial
Three serious, well-funded programmes. Three failures in a few days. None of them made the front of the week’s coverage.
The drugs that didn’t
What did make the front: Definium Therapeutics said a single dose of its LSD-based pill significantly cut symptoms of major depression in a late-stage trial, and its shares jumped 55% to a four-year high
AbbVie agreed to buy Apogee Therapeutics for $10.9 billion, its largest deal in five years, to get hold of an experimental eczema drug
Good news is real news. But notice what a week of biotech actually looks like when you count both columns: roughly as many things broke as worked.
What you’re not seeing
The failures get a one-day press release and a falling share price, then vanish. The successes get celebrated, repeated, and built into the story of an industry that keeps delivering. So the picture most of us carry — of medicine as a steady march of wins — is assembled almost entirely from the survivors.
The hard number behind this: of the drugs that make it into human testing at all, only around one in ten ever reaches approval. In cancer, it is worse. Most never finish. This is not a bad week — it is the base rate of the field, made briefly visible because three of the failures happened to be big names at once.
The system quietly reacting to it
That failure rate is also why the FDA, the US drug regulator, announced a plan this week to speed up the earliest, phase 1 trials — the first, small tests in humans — by six to twelve months, partly to keep more of that risky early work in the US rather than China
The week’s winners earned their headlines. But a clear-eyed reader holds them next to the losers nobody named — because that fuller picture, not the highlight reel, is what tells you how hard this actually is.
02 · Lesson · why it matters
Why the wins you hear about make every field look easier than it is
We judge the odds from the survivors — the ones still standing make it look like most attempts succeed, when most of them quietly failed and left.
The losers don’t send press releases
Three drugs failed this week — a Pfizer lung-cancer drug, an Exelixis colorectal drug, a Novocure brain-cancer device. You probably hadn’t heard of any of them until they fell. What you heard about instead was the LSD depression pill that worked and the $10.9 billion deal that closed.
That isn’t a fluke of one week’s news. It is how the news of every field is shaped. Success gets announced, repeated, and remembered. Failure gets a falling share price and silence. The wins stay in front of us; the losses leave the room.
You’re seeing the survivors, not the field
Here is the trap. When you want to know how likely something is to work, you look at the cases you can see. But the cases you can see have already been filtered — by who got celebrated, who got covered, who is still around to be looked at. You’re not sampling the field. You’re sampling the survivors.
In drug development the gap is enormous. Of the drugs that even reach human testing, only about one in ten is ever approved. In cancer, fewer. So nine out of ten of the bets being placed right now will fail — and almost none of those nine will ever cross your screen. The one that works will. From where you sit, the field looks like a string of wins, because the losses were edited out before you ever saw them.
The name for it
There’s a clean name for this: survivorship bias. You judge the odds from what survived, because the rest is invisible. It was first spotted on warplanes. Analysts in the war wanted to know where to add armour, so they studied the bullet holes on the planes that came back. The holes clustered on the wings and tail. The obvious move was to armour where the holes were.
It was exactly backwards. The planes they studied were the ones that made it home. The places they had no holes — the engines, the cockpit — were the places a hit was fatal. Those planes never came back to be studied. The armour belonged where the data was blank, not where it was thick. The missing cases held the answer.
Where this reaches you
This is not a story about drug companies. It is about how you read the world.
The successful founder telling you how she did it is, by definition, the one who didn’t go bankrupt — and you’ll never get the talk from the ten who ran the same playbook and lost. The supplement with the glowing reviews is showing you the people it helped, not the ones it did nothing for, who quietly stopped and moved on. The investment strategy with the great track record is often the one fund left standing out of the hundred that started. “Just keep going, it worked for me” comes only from the people for whom it worked.
You are inside this too — not above it. The wins you’ve had look, from the inside, like proof your approach is sound. But you can’t see your own near-misses the way you see other people’s, and you can’t see the people who did exactly what you did and got a worse roll of the dice. The view from any single seat is built from survivors, and yours is no exception.
What seeing the whole asks of you
Seeing the whole here is mostly about seeing the absence — learning to ask, each time someone shows you a winner, where are the ones who tried this and aren’t here to talk? The graveyard is the data. It is just harder to look at, because nothing in it is waving for your attention.
This doesn’t make the wins fake. Pfizer’s failure doesn’t make Intellia’s success less real. It means the wins are a slice, and a flattering one, and the field is the whole loaf. Hold the headline next to the ones nobody wrote, and the world stops looking like a march of triumphs and starts looking like what it is — hard, mostly unsuccessful, and worth more humility than the highlight reel invites.
03 · Lab · your turn
The Headlines You See
Estimate a field's success rate from the wins you hear about, then watch the hidden failures reveal how survivors skew the odds.
04 · Hope · carry this
The nine failures behind every approved drug aren't waste — they're how the one that works gets found, paid for by people willing to try and lose. An industry honest enough to fail in public most of the time is exactly the kind that eventually delivers the thing you'll be glad someone kept reaching for.
More from Biotech & Longevity