Regeneron Just Got Schooled by the Drug It Was Trying to Beat
Regeneron's fianlimab combo showed five extra months of progression-free survival versus Keytruda in melanoma, but missed statistical significance by a hair. Wall Street wiped 11% off the stock and analysts are calling it a worst-case scenario for the company's oncology ambitions.
The Most Expensive Near-Miss in Oncology
Imagine training for years to beat Usain Bolt in a race. You get to the finish line a step behind him, but close enough to smell his cologne. You think you won, but the clock says otherwise.
That's essentially what just happened to Regeneron. The company spent years building a melanoma combination therapy designed to dethrone Merck's Keytruda, the reigning king of checkpoint inhibitors. It ran a massive Phase 3 head-to-head trial. It got a numerical advantage. And then the statistics said: not good enough.
Wall Street responded by wiping roughly 11% off Regeneron's stock price. Analysts used words like "worst-case scenario" and "shaking our heads." One firm downgraded the stock entirely. The fianlimab dream, once projected to generate up to $1.8 billion in peak melanoma sales, has been zeroed out of multiple financial models.
What Regeneron Was Trying to Do
The drug in question is fianlimab, an antibody that blocks a protein called LAG-3 on immune cells. Think of LAG-3 as a second parking brake on your T cells (the immune cells that kill tumors). Keytruda releases the first parking brake, PD-1. Regeneron's theory: release both brakes at once, and the immune system should attack cancer more aggressively.
Regeneron paired fianlimab with its own PD-1 blocker, cemiplimab (sold as Libtayo). Two drugs versus Keytruda's one. A combination punch versus a single jab.
Early-phase trials looked extremely promising. In a smaller study of 98 advanced melanoma patients, 57% responded to the combination, with 25% achieving complete responses. That's a high bar. The Phase 3 trial was supposed to confirm those results against the gold standard.
How Close Is Close?
The primary endpoint was progression-free survival (PFS), which measures how long patients live without their cancer growing. The results, on paper, look intriguing:
High-dose fianlimab + cemiplimab: 11.5 months median PFS
Get tomorrow's biotech intelligence before your competitors.
Join thousands of biotech professionals who start their day with our free, daily briefing.
Keytruda alone: 6.4 months median PFS
That's a difference of more than five months. In cancer medicine, five months is meaningful to patients. So why is everyone calling this a failure?
Because the p-value came in at 0.0627. In clinical trials, you need to hit below 0.05 to claim statistical significance. Regeneron missed that threshold by a hair. And in drug development, a hair might as well be a mile. The FDA doesn't hand out approvals for "almost significant." Guidelines committees don't rewrite treatment protocols for "trending in the right direction."
The low-dose arm? Even worse: a p-value of 0.4661, which is statistically meaningless.
Wall Street Didn't Just Frown. It Grimaced.
The analyst reactions were swift and brutal.
Citi cut its rating from Buy to Neutral and slashed the price target from $900 to $700, arguing there are no "incremental positive catalysts" left to support a premium valuation. BMO Capital cut its target by roughly 20% and said the melanoma readout was a "defining catalyst" that failed. They explicitly said they were "shaking our heads." Evercore called it a "worst-case scenario" for investor sentiment.
RBC Capital lowered its target to $707 from $762 and tied this failure into a pattern of recent pipeline setbacks, including a lung disease drug (itepekimab) that also stumbled and ongoing issues with Eylea HD. Their warning: skeptics will now "more vocally question the company's overall direction and strategy."
Only a handful of firms maintained Buy ratings. Jefferies kept the faith, trimming its target modestly while noting the miss was "not particularly surprising" and a validation of the bear case. BofA Securities stayed constructive too, arguing that Regeneron's core earnings power (especially Dupixent, which dominates in eczema and asthma) remains intact.
Why Keytruda Is So Hard to Beat
Keytruda isn't just a drug. It's a fortress.
In melanoma alone, it generates roughly $5-6 billion annually. It's approved in both metastatic and adjuvant settings. Oncologists know it cold; they've been prescribing it for years. It has one of the deepest clinical data packages of any cancer drug ever developed.
Beating Keytruda head-to-head requires you to prove, beyond statistical doubt, that your drug is meaningfully better. And "meaningfully" has to show up in the math, not just the median. The PFS curves in Regeneron's trial apparently only separated later in follow-up, which weakens the statistical case even when the numbers look appealing on the surface.
To use a sports analogy: Keytruda is the defending champion who benefits from every close call. Challengers don't get moral victories.
The Broader LAG-3 Question
This isn't just about Regeneron. It raises questions about whether the entire LAG-3 drug class can truly deliver on its promise.
The only approved LAG-3 combination is Bristol Myers Squibb's Opdualag (relatlimab + nivolumab), which won FDA approval in 2022 for advanced melanoma. Its pivotal trial showed a median PFS of 10.1 months versus 4.6 months for nivolumab alone, a clear and statistically significant win.
But notice the comparison. Opdualag beat nivolumab monotherapy, not Keytruda. Nobody has yet run a head-to-head of Opdualag versus Keytruda. Regeneron tried the harder challenge: proving its LAG-3 combo could beat the market's preferred PD-1 drug on PD-1's home turf. It came close, but close doesn't count.
The lesson for the LAG-3 class: dual checkpoint blockade probably helps some patients meaningfully. But proving superiority over established PD-1 monotherapy in a broad, all-comers population is a monumental statistical challenge. The effect size may simply be too modest for conventional trial designs to capture cleanly.
What's Left for Regeneron's Oncology Ambitions?
Regeneron isn't abandoning the LAG-3 program entirely. Several cards remain unplayed:
Adjuvant melanoma. A separate Phase 3 trial is comparing fianlimab + cemiplimab against Keytruda in high-risk patients after surgery (preventing recurrence rather than treating metastatic disease). Different patient population, different endpoint (relapse-free survival), different math.
Head-to-head versus Opdualag. Regeneron is running a Phase 3 trial pitting its high-dose combo directly against Bristol Myers Squibb's approved LAG-3 combination. If fianlimab + cemiplimab can match or beat Opdualag, there may still be a path to market, even without beating Keytruda.
The perioperative setting. A Phase 2 trial is testing the combo before and after surgery in high-risk melanoma, a rapidly evolving area where new data could create niche opportunities.
But analysts are now deeply skeptical. BMO noted that even non-inferiority to Opdualag is no longer their base case. The fianlimab program has shifted from "flagship growth driver" to "long shot." Regeneron's near-term stock story now rests on Dupixent's continued growth, Eylea stabilization, and its Factor XI blood thinner program, which has reported Phase 2 results and moved into Phase 3 trials.
The Bigger Picture
This trial tells a story that keeps repeating in oncology. PD-1 inhibitors like Keytruda and Opdivo are so deeply embedded as standard of care that displacing them requires not just better drugs, but overwhelmingly better drugs. The bar isn't improvement; it's proof beyond reasonable doubt.
For patients, that's both good news and bad. Good, because existing treatments work well enough that they've become very hard to beat. Bad, because genuinely incremental improvements (like, say, five extra months before progression) might never get approved if the statistics don't cooperate.
Regeneron's p-value of 0.0627 will haunt the company for a while. In some alternate universe where the trial enrolled 50 more patients or the randomization broke slightly differently, that number might have crossed 0.05. But we don't live in that universe. We live in one where Keytruda remains king, and the next challenger will need to bring more than a numerical edge to claim the throne.
Funding & Financings
4 min read
Denmark's $861M Bet to Become Europe's Biotech Capital
The Novo Nordisk Foundation just made its largest donation ever: $861 million over a decade to turn Denmark's BioInnovation Institute into a European biotech powerhouse. The catch? It's not just about drugs anymore.
