Clinical trial failures

Clinical trial failures: search stopped trials and failure signals

Clinical trial failures are often hidden in registry text, status changes, and sponsor-provided stop reasons. This guide explains how to use the Clinical Trial Failures database to study terminated, suspended, and withdrawn trials and separate likely biological failure from operational or strategic stops.

What counts as a clinical trial failure?

In this database, a clinical trial failure usually means a trial record that is terminated, suspended, or withdrawn before completion. Not every stopped study failed scientifically. Some trials stop because of enrollment, funding, portfolio strategy, site operations, or regulatory decisions.

The most useful signals are the records where the stop reason suggests the intervention did not work as intended, created unacceptable safety risk, or reached futility. Those records can reveal patterns that are hard to see when reading individual registry entries one by one.

Why a searchable clinical trial failure database helps

ClinicalTrials.gov includes structured fields and free-text sponsor explanations, but the reasons behind trial stops are not always standardized. One sponsor may write lack of efficacy, another may write futility, and another may describe an endpoint or safety issue in longer language.

Clinical Trial Failures brings those records into a workflow built for scanning, filtering, comparison, and export. Researchers can move from a broad market question to a specific set of stopped trials and then verify each primary record.

How to interpret the results

Use the labels as screening signals, not final judgments. A stopped trial may have multiple causes, and registry text can be incomplete. The strongest workflow is to use the database to find candidate records, then review the original ClinicalTrials.gov entry and any related sponsor publications.

For medical and investment decisions, treat the database as research support. It is designed to reduce search time and surface patterns, not to replace primary source review.

Original dataset signals

What the current stopped-trial dataset shows

The database currently contains 23,452 stopped trial records from ClinicalTrials.gov. The useful SEO point is also the useful research point: most stopped trials are not automatically biological failures, so the page separates status from interpreted stop reason.

Stopped records23,452

Trials marked terminated, withdrawn, or suspended.

Likely biological failures1,813

Records classified as efficacy/futility or safety-driven biological failure signals.

Biological share8%

A reminder that many stopped trials are operational, strategic, or unclear.

Top stop-reason buckets

Operational
12,013
Other/unknown
9,534
Efficacy/futility
1,096
Safety
717

Largest disease areas

Oncology
7,871
Other
5,755
Infectious disease
1,700
Gastroenterology and hepatology
1,519

Example records to verify

NCT07014735

Effect of Hyperglycaemia and Moxifloxacin on QTc Interval in T2DM

Efficacy/futility

The registry stop language says the study was terminated early on futility grounds, making it a clear example of a trial-level scientific stop signal.

Open trial record
NCT05999968

Abemaciclib plus darolutamide in prostate cancer after initial treatment

Efficacy/futility

The record links termination to a related study that did not meet its primary endpoint, which is useful context for interpreting the stopped program.

Open trial record
NCT04867837

OCTAPLEX in patients with acute major bleeding on DOAC therapy

Efficacy/futility

The stop language references an interim analysis and futility based on treatment effect size, a good example of why source context matters.

Open trial record

Counts are generated from the site's current ClinicalTrials.gov-derived stopped-trial dataset. Because registry records can change, use these figures as research signals and verify important records at the source NCT page.

Frequently asked questions

Is every terminated clinical trial a failure?

No. Terminated, suspended, or withdrawn status is a starting signal. Some trials stop for non-scientific reasons such as enrollment, funding, strategy, or operations.

Where does the data come from?

The primary source is ClinicalTrials.gov registry metadata and sponsor-provided stop-reason text where available.

Can I download or filter the records?

Yes. Use the Explore page to search, filter, compare, and export stopped trial records.