Arca
Skip to content

How Drug Trials End

Most drug trials never say how they turned out. Of those that do, this shows how many met their main goal, how many missed, and how many stopped early and why, for each disease and drug.

524,548 ClinicalTrials.gov studies, each labelled by how it ended: met its primary endpoint, missed it, or stopped early and why. Most never report a result, so rates here are of the 102,310 whose ending is known.

  • Ending known: 102,310
  • Met primary endpoint: 44%
  • Stopped early: 48%

How Trials End

Trials whose ending is known.

  • Met its primary endpoint44,97844%
  • Missed its primary endpoint8,3218%
  • Stopped: lack of efficacy3,2103%
  • Stopped: safety6431%
  • Stopped: enrollment or feasibility10,97511%
  • Stopped: sponsor or business decision4,7495%
  • Stopped: funding1,7972%
  • Stopped: other or unclear27,63727%

Why Trials Stop

The reason the sponsor gave, where it fits one of these.

  • Enrollment or feasibility10,975
  • Sponsor or business decision4,749
  • Lack of efficacy3,210
  • Funding1,797
  • Safety643

Hardest Diseases

Lowest share of trials meeting their endpoint, of 300 or more with a known ending.

Most Missed Endpoints

FDA-approved drugs with the most trials that missed their primary endpoint, often in new uses.

Every Condition and Drug

Most trials first; conditions with 20 or more known endings, drugs with 10 or more

1,079 of 1,079 conditions

Breast cancer11,3258512101,41334%112
Diabetes mellitus10,7791,75718179164%61
Type 2 diabetes8,3921,5219261968%43
COVID-197,56139517691327%76
Lung cancer7,2234651961,09726%114
Depression7,13882017259452%34
Leukemia6,435608861,15233%117
Lymphoma6,428661821,03137%87
Hypertension6,35677710863851%51
Stroke6,16341310545842%33
Prostate cancer5,80936612283328%85
Non-small cell lung cancer5,61340515495527%97
Depressive disorder5,59469214450851%28
Coronary artery disease5,32354611942550%19
Heart failure5,17946315065537%49
Colorectal cancer4,76633012251034%46
Osteoarthritis4,5554077438847%26
HIV infection4,30559311331358%31
Obesity4,0713935322259%12
Postoperative pain4,0512967031743%27
Parkinson's disease3,9502753933442%29
Arthritis3,8826486938459%42
Sclerosis3,7653508041342%44
Asthma3,65850810932154%34
Hepatitis3,6266316039758%37
Chronic obstructive pulmonary disease (COPD)3,5515378728959%14
Fibrosis3,4873228037941%45
Atrial fibrillation3,4743336032746%20
Multiple myeloma3,4273393659435%59
Myocardial ischemia3,4073628224952%12
Alzheimer's disease3,2711938435231%59
Cognitive dysfunction3,2641685123037%17
Schizophrenia3,2554037930951%31
Type 1 diabetes3,2444135027656%23
Multiple sclerosis3,1663005231445%25
Hemorrhage3,1122437930039%22
Myeloid leukemia3,0763065157233%63
Anxiety disorder2,9852823817457%8
Knee osteoarthritis2,9722634221251%12
Coronary disease2,9633076520154%10
Hepatitis a2,9385315531959%28
Infarction2,89726112624341%16
Plasma cell cancer2,8793013450636%49
Obstructive lung disease2,8614097424656%9
Overweight2,7853385314363%6
Pancreatic cancer2,7761337742821%44
Melanoma2,7472276551328%57
Hepatocellular carcinoma2,7241454733428%29
Myocardial infarction2,71825111921643%10
Chronic renal insufficiency2,7142545825345%11

Outcome labels are from TrialPanorama (Apache-2.0), built from ClinicalTrials.gov records partly automatically, so some are wrong: in a sample of trials labelled stopped for lack of efficacy, about a quarter gave another reason. Met and missed come from posted results and publications, which favour trials that succeeded. Stopped-early reasons are the sponsor's own words, grouped; โ€œother or unclearโ€ is left out of the reasons chart. Conditions are MeSH terms; a trial counts under each condition and drug it names. TrialPanorama

Logos are each company's own site icon, matched by name through Wikidata.Company names and logos identify the companies in public FDA and registry records and do not imply endorsement.