Regeneron Missed by a Hair. In Oncology, That's a Mile.
Regeneron's LAG-3 combo missed its primary endpoint against Keytruda by the thinnest statistical margin, with a p-value of 0.063 versus a 0.05 threshold. Analysts have wiped up to $1.8 billion in projected peak sales from their models, and the company's oncology ambitions are looking increasingly fragile.
The Most Expensive 0.013 in Biotech
Imagine training for four years, flying to the Olympics, and losing the gold medal by six hundredths of a second. You were right there. The crowd saw you almost win. But "almost" doesn't get you on the podium.
That's what just happened to Regeneron.
The company's Phase 3 trial of fianlimab, a LAG-3 inhibitor paired with its PD-1 drug cemiplimab (Libtayo), failed to beat Merck's Keytruda in first-line advanced melanoma. The p-value landed at 0.0627. The threshold for success was 0.05. A gap of 0.013 separated Regeneron from a potential blockbuster oncology franchise and a very expensive disappointment.
Wall Street didn't care about the margin. Shares dropped roughly 12% following the announcement.
The Numbers Tell Two Stories
Here's where it gets interesting (and a little maddening for Regeneron). The drug actually looked like it was working.
Patients on the high-dose fianlimab combo had a median progression-free survival of 11.5 months. That's the amount of time before their cancer started growing again. The Keytruda group? 6.4 months. That's a numerical advantage of more than five months, which in melanoma is genuinely meaningful.
But the hazard ratio (a statistical measure of how much better one treatment is than another) came in at 0.845, with a confidence interval that crossed 1.0. Translation: the math couldn't rule out that the difference was due to chance. The benefit seemed to appear late, with the survival curves separating only after patients had been on treatment for a while. That late separation killed the pre-planned statistical analysis.
The low-dose combo fared worse. Its hazard ratio was 0.931 versus Keytruda, with a p-value of 0.47. That's not close. That's not even in the same zip code.
Why "Close" Doesn't Count in Drug Development
In everyday life, we celebrate close calls. In clinical trials, they're brutal.
Get tomorrow's biotech intelligence before your competitors.
Join thousands of biotech professionals who start their day with our free, daily briefing.
Regulators need statistically significant proof that a drug works before they'll approve it. Doctors need that proof before they'll switch patients off a treatment that already has years of real-world data behind it. And Keytruda isn't just any drug; it's the best-selling cancer therapy on the planet, deeply embedded in oncology practice across dozens of tumor types.
Trying to dethrone Keytruda head-to-head is like challenging the reigning champion on their home court. You don't just need to be good. You need to be provably better. Regeneron was good. It wasn't provably better.
No new safety signals emerged, which is a small consolation. The combo was well-tolerated. But tolerability without statistical significance is like a restaurant with great ambiance and mediocre food: it's not enough.
A Pattern That's Getting Hard to Ignore
This isn't Regeneron's first high-profile miss. The company previously pulled fianlimab out of lung cancer development after Phase 2 data didn't justify moving to Phase 3. Its IL-33 inhibitor itepekimab also stumbled in a Phase 3 COPD trial. And the blood cancer program around odronextamab has faced its own regulatory headwinds.
Analysts widely treated the melanoma readout as a key catalyst for the first half of 2026. It defined things, all right: just not the way Regeneron wanted.
Both firms have now removed fianlimab from their financial models for melanoma. BMO had previously estimated risk-adjusted peak sales of roughly $1.8 billion in this indication. Those numbers are now zero in their base cases.
The Biology Problem Behind the Business Problem
Zoom out from Regeneron for a moment, because this story is bigger than one company.
LAG-3 is a protein on the surface of T cells (your immune system's soldiers). It acts like a brake, telling T cells to calm down. In cancer, that brake helps tumors hide from the immune system. The idea behind LAG-3 inhibitors is simple: release the brake, let T cells attack the tumor.
The problem? LAG-3 rarely works alone. Exhausted T cells in tumors typically have multiple brakes engaged simultaneously, including PD-1, TIM-3, and TIGIT. Blocking just one brake often isn't enough. That's why the entire LAG-3 field has shifted toward combination approaches, pairing LAG-3 blockers with PD-1 inhibitors to release two brakes at once.
Bristol-Myers Squibb proved this concept with Opdualag (relatlimab plus nivolumab), which showed significantly improved progression-free survival in melanoma compared to nivolumab alone. But here's the key distinction: BMS compared its combo against PD-1 monotherapy. Regeneron tried to beat PD-1 monotherapy (Keytruda) with a different PD-1 plus LAG-3 combo. The bar was essentially the same, but Regeneron's statistical plan didn't hold up.
What's Left in the Tank
Regeneron isn't abandoning fianlimab entirely. A separate Phase 3 trial comparing the high-dose fianlimab/cemiplimab combo directly against BMS's Opdualag is still running. If that study hits its endpoints (overall survival or objective response rate), it could carve out a niche for fianlimab as a competitive alternative to the only approved LAG-3 combo.
But expectations have been reset dramatically. Analysts describe a win against Opdualag as "possible but not base case." In investor-speak, that translates to: we'll believe it when we see it.
There's also the Harmony adjuvant trial, testing fianlimab plus cemiplimab in patients whose melanoma has been surgically removed but who are at high risk of recurrence. Different setting, different endpoints, different shot on goal.
Where Regeneron Goes From Here
The good news for Regeneron shareholders: this isn't an existential crisis. The company still has Dupixent (the blockbuster allergy and asthma drug), Eylea (a leading eye treatment), and a promising obesity program. Its Factor XI antibody program in thrombosis is generating real excitement, with pivotal data expected around 2027.
But the oncology narrative has taken a genuine hit. Two years ago, Regeneron was pitching a story about becoming a major immuno-oncology player, with LAG-3 as a cornerstone. That cornerstone just cracked.
The company can still be a successful biotech giant. It just probably won't become an oncology giant through LAG-3 alone. Sometimes the most important thing a clinical trial tells you isn't in the data; it's in what the data forces you to rethink.
Deals & M&A
4 min read
The Italian Family That Just Spent $4.1 Billion to Break Into America
A 107-year-old Italian family pharma just dropped $4.1 billion on a Florida rare disease company, paying roughly three times its own annual revenue. The deal gives Angelini instant access to the U.S. market and a portfolio generating nearly $600 million a year.
