The TheraRadar Brief
Every drug has more than one story. Most never get told.
When an IBD Trial Is Terminated, the Drug Usually Didn’t Fail
Of 214 industry IBD trials terminated or withdrawn over three decades, only ~23% stopped for efficacy or safety. The rest ran out of patients, money, or corporate priority — the trial failed, not the biology.
Get the next TheraRadar Brief in your inbox
Drug development, biology, and market dynamics — free, every week.
A terminated trial reads like a failed drug. Usually it isn’t. Across 214 stopped industry IBD trials, the single most common reason is a business or portfolio decision — and only about 23% were halted because the drug looked unsafe or ineffective. The trials that did stop for efficacy had run to full size; the operational stops had barely begun.
Over three decades, industry has run about a thousand interventional IBD trials — a trickle in the early 2000s, climbing to roughly fifty a year now. This brief is about the ones that stop early.
Industry-sponsored interventional IBD trials registered per year
Does a halted trial mean the drug failed?
When a trial is terminated, the natural reading is that the drug failed — but that default is wrong more often than not. A trial can be stopped because it couldn’t recruit patients, because the sponsor was acquired or reprioritised, or because a manufacturing lot fell through — none of which says anything about whether the molecule works. So we measure how often a termination is actually about the drug, using what the sponsors themselves said.
ClinicalTrials.gov records a free-text Why Study Stopped (whyStopped) note on every terminated or
withdrawn trial — the sponsor’s own short account of what happened. Reading that note is the whole method
here: it is how we recover the actual reason a trial stopped, instead of assuming the drug
failed. We took all 214 industry IBD trials that were stopped
(182 terminated, 32 withdrawn — from 1,007
industry interventional trials in scope) and classified each note into one of seven categories: lack of
efficacy, safety, enrollment, business/portfolio, operational/supply, other, or unstated.
The named cases
Start with the notes themselves. A few, verbatim — each auditable against the trial record:
| Efficacy | mongersen (SMAD7 antisense, Celgene) NCT02641392 | “…terminated by Sponsor following a recommendation from external DMC based on lack of emerging efficacy.” |
| Safety | brodalumab (anti-IL-17RA) NCT01150890 | “…terminated early based on an imbalance in worsening Crohn’s disease in active treatment.” |
| Business | brazikumab (anti-IL-23, AstraZeneca) NCT03961815 | “Strategic decision to discontinue the development of brazikumab in inflammatory bowel disease.” |
| Business | bertilimumab (anti-eotaxin-1) NCT01671956 | “Sponsor Decision.” |
| Enrollment | TAK-018 (Takeda) NCT03943446 | “…discontinue study due to inability to recruit the expected number of subjects.” |
mongersen and brodalumab are clean reads on the biology — the SMAD7 Phase-3 futility failure, and the IL-17 antibody that worsened Crohn’s. brazikumab’s “strategic decision” shows the other side: a business label can hide an efficacy concern.
Why are IBD trials terminated early?
Business and portfolio decisions are the largest single category at 36% — a drug shelved in a reshuffle, a merger, a strategic exit. Inability to enroll is next (22%). Lack of efficacy and safety — the two reasons that are actually about the drug — together account for just 50 of the 214 stops, about 23%. Put the other way: when an IBD trial is pulled early, roughly 77% of the time the reason says nothing about whether the intervention worked. A caveat on direction: a “strategic” label can mask an efficacy concern (see brazikumab above), which biases the efficacy count down — so ~23% is a ceiling, not a floor.
| Stated reason | Total | Crohn’s | UC | Both / IBD |
|---|---|---|---|---|
| Business / portfolio | 77 | 35 | 33 | 9 |
| Couldn’t enroll | 47 | 24 | 15 | 8 |
| Lack of efficacy · about the drug | 42 | 13 | 27 | 2 |
| Unstated | 28 | 16 | 8 | 4 |
| Safety · about the drug | 8 | 2 | 4 | 2 |
| Other | 8 | 3 | 3 | 2 |
| Operational / supply | 4 | 1 | 2 | 1 |
| All stops | 214 | 94 | 92 | 28 |
The split exposes an asymmetry the totals hide: efficacy stops skew to ulcerative colitis — 27 of the 42 are UC versus 13 in Crohn’s. When an IBD drug is halted because it didn’t work, it is more often a UC program.
The size tell
The stated reasons line up with how far each trial got. Trials stopped for enrollment had barely opened; trials stopped for efficacy had run a full-sized study and then the drug lost.
Why it matters
By reading terminations as drug failures, we misjudge the field twice over. We overstate how often the biology has actually been tested and lost — most early stops are feasibility and portfolio problems, a lesson about trial design and operational risk, not target choice.
It helps to step back from individual trials to whole programs — a program being one drug’s full development effort in IBD, which usually spans several trials.
One caveat keeps this honest. Those 214 terminations map to 107 distinct programs — the loud failures. They are not the whole story. A larger group of programs (192) quietly completed a trial and then never advanced, leaving no termination to count at all. These silent failures are the bigger problem: a drug that runs a full study, posts an unremarkable result, and is shelved leaves no “terminated” flag — so it never appears in a termination census like this one. Counting only the loud failures both understates how often IBD programs die and hides where they die.
Need the full failure map for your pipeline?
Lessons from the Failures scores all 437 novel industry IBD programs — loud and silent — by mechanism, indication, and region, showing where IBD development actually dies. We license the full report and run custom attrition analyses on your own targets.
Common questions
Does a terminated trial mean the drug failed?
Usually not. Of 214 stopped industry IBD trials, only about 23% stopped for efficacy or safety. The other ~77% stopped for business decisions, enrollment, operations, or reasons the sponsor never stated.
What most often stops an IBD trial early?
Business and portfolio decisions, at 36% — the single largest category. Inability to enroll is next (22%). Lack of efficacy and safety together are only 50 of 214 stops (~23%).
Terminated vs. withdrawn — what’s the difference here?
Both are early stops on ClinicalTrials.gov. Terminated means the study stopped early after it had begun enrolling participants — anyone enrolled is no longer being examined or treated. Withdrawn means it stopped early before enrolling its first participant.
How do you recover the real reason a trial stopped?
Read the sponsor’s free-text whyStopped note — its own short account of what happened. Every label here is auditable against clinicaltrials.gov/study/<NCT>.
Sources & method
All counts are computed from a frozen ClinicalTrials.gov snapshot
(2026-06-15) by released code, and reproduce the Pro analysis exactly. Cohort: condition = inflammatory
bowel disease / Crohn’s / ulcerative colitis, studyType = INTERVENTIONAL, lead sponsor
INDUSTRY, status TERMINATED or WITHDRAWN (214 trials). Each reason is the
trial’s verbatim whyStopped field, classified by a keyword baseline and a language model over
a fixed codebook (efficacy · safety · enrollment · business · operational · other · unstated); every
label is auditable against clinicaltrials.gov/study/<NCT>. This is an analytical brief,
not medical advice.
Related TheraRadar reading: TL1A: The Target Behind the $18 Billion Bet — the IBD target everyone is now chasing. Entyvio Doesn’t Go to Your Brain — the one gut-selective IBD drug, and the challengers that failed trying to copy it.
Go deeper with Pro
Pro gives you the data — patent cliffs, trial analytics, competitive landscapes, revenue tracking. Briefs tell the story with the data. Launch pricing: $99/month.
See Pro plansGet the next Brief in your inbox — free
Every drug has multiple stories. Most never get told. One brief every week. No spam.
Archive lives at /briefs/. Unsubscribe anytime.