J&J's Antidepressant Keeps Striking Out, and Nobody's Surprised
J&J and Contineum's rapid-acting depression drug just failed its second Phase 2 trial, and it's part of a brutal streak for novel antidepressant mechanisms. Wall Street barely flinched, but the implications for beyond-SSRI psychiatry are worth paying attention to.
Depression affects roughly 280 million people worldwide. You'd think that would make it easy to build a better antidepressant. You'd be wrong.
Johnson & Johnson and its partner Contineum Therapeutics just watched their experimental depression drug, JNJ-5120 (also known as PIPE-307), fail a Phase 2 trial. The drug didn't beat a sugar pill at reducing depression symptoms. And the worst part? This isn't even the first time this molecule has disappointed.
A Drug That Works Fast (In Theory)
To understand why anyone cared about this drug in the first place, you need to know the biggest complaint about existing antidepressants: they're slow. SSRIs, the most commonly prescribed class, can take four to six weeks to kick in. If you're in a dark place, a month feels like a lifetime.
JNJ-5120 was supposed to be different. It targets something called the M1 muscarinic receptor, a protein on the surface of brain cells that helps regulate signaling between neurons. The idea is that by selectively blocking this receptor, you can essentially flip a switch in the brain's prefrontal cortex: calming down inhibitory signals, boosting excitatory ones, and triggering the growth of new neural connections. Think of it like unclogging a drain so that the good neurochemistry can flow again.
The promise? Antidepressant effects within days, not weeks. The science behind it wasn't pulled from thin air, either. An older drug called scopolamine had already shown that blocking M1 receptors could produce rapid mood improvements, giving researchers reason to believe the target was "clinically validated."
So Contineum built a cleaner, more selective version. J&J liked it enough to license it. And then they tested it in humans.
The MOONLIGHT Trial Goes Dark
The study was called MOONLIGHT-1, and it had all the hallmarks of a well-designed proof-of-concept trial: randomized, double-blind, placebo-controlled, conducted across multiple sites. It enrolled 107 adults with major depressive disorder and tested JNJ-5120 as a standalone treatment.
Get tomorrow's biotech intelligence before your competitors.
Join thousands of biotech professionals who start their day with our free, daily briefing.
The primary goal was straightforward: show that patients on the drug improved more than patients on placebo, as measured by the MADRS score (a standard depression rating scale) at Day 5. That tight timeline wasn't arbitrary; it was the whole point. If this drug was going to be a rapid-acting antidepressant, it needed to show results fast.
It didn't. The drug failed to separate from placebo on the primary endpoint. Patients taking the real drug didn't improve meaningfully more than those taking a dummy pill.
The silver lining, if you can call it that: the drug was well-tolerated with no new safety signals. So the molecule isn't dangerous. It just doesn't appear to work for depression.
Déjà Vu, All Over Again
If this story feels familiar, it should. JNJ-5120 had already stumbled before, in a separate Phase 2 study testing it for multiple sclerosis. That trial also failed to show benefit.
Two swings in two different diseases. Two misses. At some point, you have to ask whether the bat is the problem.
J&J says it's still reviewing the full dataset, including "prespecified exploratory endpoints," to figure out next steps. That's the clinical trial equivalent of saying "we're going through the game tape." It leaves the door cracked open, but nobody should expect a triumphant return without some very creative data reinterpretation.
The $75 Million Bet
Let's rewind to April 2023, when this partnership was announced. Contineum granted J&J an exclusive worldwide license to develop and commercialize PIPE-307 across all indications. The price tag: $50 million upfront in cash, plus a $25 million equity investment from J&J's venture arm. On top of that, Contineum stood to earn more than $1 billion in milestone payments and royalties ranging from the low-double digits to the high teens on any net sales.
That's a substantial bet on a single molecule. For Contineum, the structure was smart: they got $75 million in the door and handed most of the development risk to one of the world's biggest pharma companies. For J&J, the deal was supposed to bolster a CNS (central nervous system) pipeline that has been a mix of big wins and quiet disappointments.
Now, with two failed Phase 2 trials on the books, those milestone payments look increasingly theoretical.
Wall Street Shrugged
You might expect a Phase 2 failure to crater Contineum's stock. It didn't. Shares dipped about 8% in after-hours trading, which for biotech is basically a rounding error.
The reason? Analysts had already priced in the risk. William Blair noted they'd viewed the study as a long shot and had previously estimated about a 35% probability of success, which tells you the Street wasn't exactly holding its breath. After the Phase 2 readout, they removed JNJ-5120 from their valuation model.
More importantly, Contineum's real value story lives elsewhere. The company's lead asset is PIPE-791, a different molecule targeting a different mechanism. William Blair explicitly said there's "little read-through" from the PIPE-307 failure in MS to the MDD program. In other words: the market is betting on a different horse from the same stable.
Depression Trials: Where Good Mechanisms Go to Die
Zoom out, and this failure fits a depressingly (pun intended) familiar pattern. The 2024 to 2026 stretch has been brutal for novel antidepressant mechanisms beyond traditional SSRIs. The graveyard is getting crowded.
NBI-1070770 for major depressive disorder? Failed its Phase 2 primary endpoint. SPN-820 for treatment-resistant depression? Couldn't beat placebo. BHV-7000? Missed on MADRS scores. BI 1358894, a TRPC4/5 channel inhibitor? Whiffed on both primary and secondary endpoints.
The list goes on. Glutamate-based approaches, NMDA receptor strategies, potassium channel openers: the mechanisms are creative, the preclinical data look compelling, and then the drugs face real patients in real trials and fall flat.
Why does this keep happening? A few reasons keep coming up:
The placebo problem is enormous. Depression trials are notorious for high placebo response rates. When a significant chunk of patients improve just because they're getting attention and care in a clinical setting, it becomes incredibly hard for an experimental drug to show a meaningful advantage. It's like trying to prove you're a great cook when everyone already loves the free bread.
Depression isn't one disease. "Major depressive disorder" is a broad umbrella covering patients with wildly different underlying biology. A drug might genuinely help a specific subgroup, but that signal gets drowned out when you enroll a heterogeneous mix of patients. Imagine testing a flu drug on everyone who walks into a clinic with a cough; some have the flu, some have allergies, and some just swallowed wrong.
Translating fast-acting biology into real-world benefit is unreliable. Scopolamine's rapid antidepressant effects inspired this whole class of research. But recreating that magic in a selective, well-tolerated pill has proven elusive. The brain is not a simple machine, and targeting one receptor cleanly doesn't guarantee the downstream cascade of effects will produce the desired outcome.
J&J's Bigger Picture: Bruised but Not Broken
For J&J, this setback is real but manageable. The company's neuroscience strategy has been reshaped around two commercial pillars: Spravato (esketamine), its rapid-acting nasal spray for treatment-resistant depression, and Caplyta (lumateperone), which it acquired through the $14.6 billion purchase of Intra-Cellular Therapies.
Spravato remains a growing franchise, with J&J investing in lifecycle studies covering anhedonia and suicidal ideation. Caplyta is approved for both schizophrenia and adjunctive treatment of major depressive disorder, giving J&J a two-track commercial business in psychiatry.
The company has reportedly trimmed some earlier-stage CNS programs in 2026, sharpening its portfolio around the most promising assets rather than placing a broad array of exploratory bets. Losing JNJ-5120 hurts, but it doesn't fundamentally alter J&J's neuroscience thesis. The company has other cards to play.
What Comes Next
J&J says it's reviewing the exploratory endpoints from MOONLIGHT-1. In biotech, that's often the last chapter before a quiet discontinuation. Sometimes companies find a subgroup that responded, or a secondary measure that showed a signal, and use that to justify one more study. But with two consecutive Phase 2 failures across different indications, the bar for continuing development is high.
For Contineum, the financial damage is limited. J&J bore the development costs, and the company's pipeline story hinges on PIPE-791. If anything, this failure serves as a reminder of why Contineum's licensing strategy was clever: let the big pharma partner absorb the clinical risk while you collect upfront payments and focus your internal resources elsewhere.
For the broader field of depression research, this is another data point in an increasingly clear trend. Moving beyond SSRIs is one of the hardest problems in psychiatry. The biology is tantalizing, the unmet need is massive, and the clinical trials keep delivering the same answer: not yet.
Somewhere out there, the next generation of antidepressants is waiting to be discovered. But after watching another promising mechanism stumble in Phase 2, you'd be forgiven for feeling a little, well, depressed about the timeline.
Clinical & Regulatory
5 min read
The FDA Just Opened the Door to Prescribing Magic Mushrooms
The FDA held a landmark public hearing on psychedelic therapy access and safety, and it wasn't asking *if* these drugs should be approved. It was asking *how*. With Compass Pathways' psilocybin potentially months from a historic first approval, the real debate is just getting started.
