BMS's Lung Scarring Drug Passed Its Biggest Test. Then Things Got Complicated.
BMS's admilparant just became the first LPA1 antagonist to hit a Phase 3 endpoint in IPF, validating a brand-new mechanism for a brutal lung disease. But a high-dose miss and a patient death are complicating what should have been a clean win.
The Good News Came With an Asterisk
Idiopathic pulmonary fibrosis is one of the cruelest diagnoses in medicine. Your lungs slowly turn to scar tissue, like a sponge hardening into concrete. Breathing gets harder every year. And the drugs we have? They slow the damage. They don't stop it. They definitely don't reverse it.
So when Bristol Myers Squibb announced that its experimental drug admilparant hit its primary endpoint in a Phase 3 IPF trial, that should have been a champagne-popping moment. A potential new treatment for a disease that desperately needs one.
But this story has layers. The lower dose worked. The higher dose didn't. And a patient died. Welcome to the messy reality of drug development.
One Dose Up, One Dose Down
The ALOFT-IPF trial tested admilparant at two doses (60 mg and 120 mg) against placebo. The goal: slow the decline in FVC, which is basically a measure of how much air your lungs can hold. Think of it as a report card for lung function. Over 52 weeks, the drug needed to prove it could protect that capacity better than a sugar pill.
At the 60 mg dose, admilparant delivered. Patients saw a statistically significant preservation of lung function compared to placebo. That's a genuine win in a disease where even modest improvements matter.
The 120 mg dose, though? It slowed lung function decline numerically, but didn't hit statistical significance. In drug development, "close but not significant" is like finishing second in a playoff game: nobody remembers you fondly.
This creates an awkward paradox. In pharmacology, you generally expect a bigger dose to produce a bigger effect. That's the dose-response relationship, and it's one of the first things regulators look for when evaluating whether a drug truly works. When the higher dose whiffs while the lower dose connects, it raises eyebrows.
The Safety Shadow
Then there's the death.
BMS disclosed that a patient in the 120 mg arm died of multi-organ failure following acute liver injury. An independent adjudication committee judged the death to the study drug, and an independent safety monitoring board recommended the trial continue.
Get tomorrow's biotech intelligence before your competitors.
Join thousands of biotech professionals who start their day with our free, daily briefing.
"unlikely related"
That's about as reassuring as these things get, but "unlikely related" isn't the same as "unrelated." BMS also amended the trial protocol to add liver injury as an important potential risk, which tells you the company is taking the signal seriously even if they don't think it's a direct drug effect.
To be fair, the overall safety profile looked solid. Adverse event discontinuations were low across both doses and comparable to placebo. BMS described admilparant as "well tolerated." But that single death, combined with the high-dose miss, creates a narrative problem that's hard to shake.
Why This Drug Matters Anyway
Let's zoom out. IPF treatment in 2026 still revolves around two aging workhorses: nintedanib and pirfenidone. Both slow lung scarring, but neither stops it, and both come with side effects that many patients struggle to tolerate long-term. A newer option, nerandomilast, was approved in 2025, but the fundamental unmet need hasn't changed. Patients are still losing ground. The disease still wins eventually.
Admilparant attacks the problem differently. It's an LPA1 antagonist, which means it blocks a receptor that acts like a "come here" signal for fibroblasts (the cells that build scar tissue). Imagine fibroblasts as overeager construction workers showing up to your lungs uninvited, laying down collagen like they're pouring concrete. LPA1 is the foreman calling them to the job site. Block the foreman, and you slow down the whole operation.
This mechanism targets the problem upstream, before the scar tissue gets laid down. That's different from existing drugs, which largely work further downstream. If admilparant's Phase 3 success holds up under full data scrutiny, it would be the first late-stage validation of LPA1 antagonism in IPF, potentially opening a whole new therapeutic approach.
What Happens Next
BMS hasn't released the granular data yet: no exact effect sizes, no p-values, no confidence intervals. Those details will come at a future medical meeting, and they'll matter enormously. The difference between "low dose clearly worked" and "low dose barely squeaked by" could reshape the entire filing strategy.
The regulatory path forward has a few likely beats:
The 60 mg dose becomes the filing dose. With a clean primary endpoint hit and a favorable safety profile, this is the obvious play. BMS will almost certainly center its regulatory submission around the lower dose.
Regulators will scrutinize the liver signal. Even with the death adjudicated as unlikely related, expect the FDA to dig deep into hepatic monitoring data. Labeling requirements, mandatory liver function testing, and post-marketing commitments are all on the table.
The dose-response gap will be debated. BMS will need a convincing explanation for why 120 mg didn't outperform 60 mg. Possible reasons include pharmacokinetic ceiling effects, trial design factors, or simple statistical noise. Without a compelling answer, regulators may view the overall evidence with caution.
There's also a potential reinforcement coming. BMS has a second Phase 3 trial, ALOFT-PPF, fully enrolled in patients with progressive pulmonary fibrosis (a broader population). Results are expected in early 2027. A strong showing there could bolster the overall data package and smooth the regulatory conversation.
The Competitive Picture
Admilparant isn't alone in the LPA1 space, but it's clearly the frontrunner. The closest competitor is PIPE-791 from Contineum Therapeutics, which is running a Phase 2 trial (called PROPEL-IPF) in about 324 patients. That trial is smaller, earlier, and won't produce data that directly challenges admilparant's Phase 3 results anytime soon.
BMS's earlier LPA1 program, BMS-986020, actually proved the concept years ago but was shelved because of liver toxicity specific to that compound. Admilparant was designed as the next-generation version, engineered to avoid that liability. The irony of a liver-related death showing up in the new program, however unlikely the connection, won't be lost on anyone.
The Bottom Line
BMS has a genuine proof-of-concept win in one of the hardest indications in medicine. The 60 mg dose delivered, the mechanism looks promising, and IPF patients desperately need new options. But the high-dose miss muddies the dose-response story, the liver safety signal will invite regulatory scrutiny, and we still haven't seen the full dataset.
It's the biotech equivalent of acing the interview but stumbling on the follow-up questions. You'll probably get the job. But the hiring committee has notes.
Clinical & Regulatory4 min read
A $22 Billion Bet Just Hit Its First Big Speed Bump
Daiichi Sankyo and Merck pulled their FDA application for an antibody-drug conjugate in small cell lung cancer before regulators could reject it. It's the first real stumble for a $22 billion ADC partnership that was supposed to be a sure thing.
