Endpoint failure evidence

Failed endpoint clinical trials: search endpoint and futility signals

Failed endpoint clinical trials are among the most useful records for understanding weak clinical evidence, but they need careful interpretation. This page focuses on stopped records where the source language points to endpoints, futility, lack of efficacy, insufficient benefit, or treatment-effect concerns.

Laptop showing an analytics dashboard for reviewing research data

What failed endpoint language usually means

A failed endpoint signal usually means the registry language says the study did not show enough evidence on a planned clinical objective, or that continuing the trial was unlikely to demonstrate the intended treatment effect.

The wording can vary. Some records mention futility directly. Others mention lack of efficacy, failed endpoints, insufficient benefit, no treatment effect, or no survival benefit. The database groups these records so they can be searched without relying on one exact phrase.

Why endpoint records are high-value for research

Endpoint-related stops are closer to the core clinical question than many other stopped-trial records. They can help users identify weak efficacy patterns by disease area, phase, sponsor, intervention, or modality.

They are also easy to overread. A failed endpoint in one patient population does not prove a mechanism is dead everywhere. Endpoint choice, dose, comparator, trial size, and patient selection all matter.

How to use this page

Use this page to understand the dataset slice, then open Explore to inspect individual records. The strongest workflow is to filter by efficacy/futility, add a disease area or sponsor, and then verify the NCT source record.

For any important record, check the registered endpoints, enrollment, phase, update dates, sponsor language, publications, and regulatory materials before drawing a conclusion.

Original dataset signals

Endpoint and weak-efficacy signals in the dataset

This page uses a targeted slice of the database: stopped records with endpoint, efficacy, futility, failed-to-meet, lack-of-benefit, or treatment-effect language. That makes the page stronger than a generic definition because it is anchored in source-derived records.

Endpoint-related records1,427

Stopped records matching endpoint, efficacy, futility, or weak-benefit language.

Terminated records1,358

Most records in this slice have TERMINATED status.

Largest disease area422 Oncology

Oncology is the largest endpoint-related disease-area slice.

Endpoint-related status mix

Terminated
1,358
Withdrawn
49
Suspended
20

Largest disease areas

Oncology
422
Other
320
Neurology
109
Infectious Disease
108

Example records to verify

NCT05491317

Immunoradiotherapy combinations in metastatic solid tumors

Efficacy/futility

The sponsor decided not to proceed to randomized Phase 2 due to lack of efficacy, making this a concrete endpoint-adjacent failure signal.

Open trial record
NCT04910269

Outpatient treatment with anti-coronavirus immunoglobulin

Efficacy/futility

The registry stop reason says the study stopped for futility, with no safety issues noted in the short source language.

Open trial record
NCT04173273

Oral etrasimod in moderately to severely active Crohn's disease

Efficacy/futility

The registry language says the study was discontinued due to lack of efficacy in a sub-study.

Open trial record

Counts are generated from the current ClinicalTrials.gov-derived stopped-trial dataset. Endpoint signals are screening labels and should be verified against the primary NCT record, endpoints, protocol context, and sponsor disclosures.

Frequently asked questions

Does a failed endpoint prove a drug failed?

No. A failed endpoint is a strong signal, but interpretation depends on endpoint design, population, dose, comparator, phase, and available source documents.

What search terms help find endpoint failures?

Useful phrases include failed endpoint, lack of efficacy, futility, failed to meet, insufficient benefit, treatment effect, and no survival benefit.

Can I filter endpoint signals by sponsor?

Yes. Open Explore and combine efficacy/futility or endpoint keywords with sponsor, disease area, phase, and status filters.