The Fibrosis Drug That Killed a Patient Nobody Can Identify
A patient death in Bristol Myers Squibb's phase 3 fibrosis trial has shaken confidence in admilparant and the broader LPA1 drug class. The catch: the data are still blinded, and nobody can confirm whether the drug was even responsible.
A Death With No Answers
Somewhere inside a blinded clinical trial, a patient died. Their organs failed. Their liver was wrecked. Their blood cells were attacked by their own immune system. And nobody outside Bristol Myers Squibb can tell you whether the experimental drug they were testing actually caused it.
That's the unsettling reality facing BMS's admilparant program right now. The company disclosed that a "limited number" of liver-related safety events, including one fatal case, emerged during the phase 3 program for this fibrosis drug. Because the trial data are still blinded, it's not even confirmed whether the patient who died was taking admilparant or a placebo.
Welcome to one of the messiest situations in drug development: a safety signal that's too serious to ignore and too ambiguous to act on.
What Actually Happened
Let's rewind. Back in May 2026, BMS quietly amended the protocol for its EU phase 3 long-term extension study. The changes added closer liver-function monitoring and more formal stopping rules for possible drug-induced liver injury. The company also designated liver injury as an "important potential risk."
That kind of mid-trial protocol change is like a restaurant quietly updating its menu to warn about allergens. Sometimes it's precautionary. Sometimes someone already got sick.
In this case, BMS said most of the liver abnormalities they'd seen were mild, transient, or consistent with hepatic adaptation (the liver's version of adjusting to a new roommate). An independent safety monitor reviewed the data multiple times and recommended the study continue.
But then there's the death. One report described it as multiorgan failure with liver damage, immune-mediated destruction of red blood cells, and widespread clotting. Another described it as acute hepatic injury with liver cell death and autoimmune hemolytic anemia. Those are two slightly different framings of what sounds like a catastrophic cascade.
The Drug, the Target, and the Ghosts of Programs Past
Get tomorrow's biotech intelligence before your competitors.
Join thousands of biotech professionals who start their day with our free, daily briefing.
Admilparant works by blocking something called the LPA1 receptor. Think of this receptor as a doorbell that, when rung by a molecule called lysophosphatidic acid, tells cells to start building scar tissue. In diseases like idiopathic pulmonary fibrosis (IPF), that scarring slowly suffocates the lungs. Admilparant essentially disconnects the doorbell.
The phase 3 program is testing admilparant in both IPF and progressive pulmonary fibrosis (PPF), with the primary goal of showing it slows the decline in lung function over 52 weeks, measured by changes in forced vital capacity (FVC), which is basically how much air your lungs can push out.
Here's what makes this liver signal especially nerve-wracking: this drug class has been here before. A previous BMS compound targeting the same receptor, called BMS-986020, was abandoned after a phase 2 trial turned up elevated liver enzymes, abnormal bile acids, and gallbladder inflammation. At the time, follow-up studies suggested the toxicity was specific to that particular molecule, not an inherent problem with blocking LPA1.
Admilparant was supposed to be the redemption arc. The cleaner, safer successor. This latest disclosure raises an uncomfortable question: what if the problem isn't the molecule, but the target?
Wall Street's Verdict: Overreaction or Early Warning?
BMY shares took a brief hit on the news, and analysts were split on how worried investors should be.
Wells Fargo acknowledged that "hepatic safety is now a meaningful overhang," while cautioning it was still unclear whether this was a drug effect or background noise from the patient population.
RBC Capital Markets analyst Brian Abrahams said the news would likely "reaffirm historical liver-toxicity concerns" for the LPA1 class, but noted there were too many variables to call it a true safety signal just yet.
Leerink Partners analyst Roanna Ruiz went further, suggesting it may not be a drug-related issue at all, much less a class-wide LPA1 problem.
And UBS pointed out that adding routine liver tests mid-study is actually fairly common in IPF trials, and that the fatal case had unclear causality.
The consensus, if you can call it that: the initial sell-off was probably too severe given the evidence, but the overhang is real. Any confirmed hepatic signal could slow or complicate the program significantly.
The Ripple Effect Nobody's Talking About
Admilparant isn't the only LPA1 antagonist in the fibrosis space. Contineum Therapeutics is developing PIPE-791, another drug targeting the same receptor. If admilparant validates the class with clean efficacy and manageable side effects, competitors benefit. If it flags a liver problem baked into the biology of LPA1 antagonism, the entire field takes a hit.
That's the thing about drug classes: they're like restaurant chains. One food poisoning scandal at a single location, and suddenly nobody wants to eat at any of them.
The FDA's typical playbook for drug-induced liver injury (DILI) in clinical trials involves enhanced monitoring, expert hepatology review, and sometimes external adjudication of cases. A single fatality doesn't automatically kill a program, especially when causality is unestablished. But the regulatory bar gets higher, the scrutiny gets sharper, and the path to approval gets longer.
What Comes Next
Several critical unknowns remain. Will BMS pause, modify, or continue the program beyond the monitoring changes already in place? Will regulators request additional data or a formal review? And perhaps most important: when the phase 3 results eventually read out, will the efficacy be strong enough to justify whatever risk the liver signal represents?
In fibrosis, the benefit-risk calculation is especially fraught. IPF is a progressive, fatal disease with limited treatment options. Patients and their doctors are often willing to accept more risk if a drug genuinely slows the disease. But "willing to accept more risk" has limits, and a fatal liver event (even an unexplained one) tests those limits hard.
For now, the admilparant story sits in an uncomfortable gray zone. Not a failure, not a vindication, not even a clear data point. Just a death, a blinded trial, and a lot of questions that won't be answered until the data are unblinded and the phase 3 results come in.
The only thing anyone can say for certain is that this program just got a lot more interesting, and a lot more stressful, for everyone involved.
4 min read
The Pill That Could Replace the Needle in Breast Cancer Treatment
Genentech's oral pill giredestrant just posted Phase 3 results in the NEJM that could shake up breast cancer treatment. In patients whose cancers outsmarted standard therapy, it cut the risk of progression by up to 62%, and Roche already has an FDA application under review.
