Biology Atlas

Map / Concepts

concept · intermediate · partly written

Animal-to-human translation figures

The published translation figures count different things; this record gives each with its definition and denominator.

Also called: translation rate, animal model concordance

The short answer

No single translation rate exists. Each published figure counts a different starting pool, so figures cannot be averaged or compared without their denominators. 123

Reported, not yet replicated

Figures and what they count more

37% (28 of 76; 95% CI 26 to 48%) of animal studies in high-impact journals (cited over 500 times, 1980 to 2000) replicated in a human RCT; 18% were contradicted and 45% untested; the sample had no negative study. 27 of 101 high-impact basic-science articles (1979 to 1983) reached an RCT, 19 a positive one and 5 licensed. 50% to any human study, 40% to an RCT and 5% to approval for the 281 therapies with at least 10 years since the first animal study (45%, 36% and 4% across all 367; median lags 5, 7 and 10 years). 71% of human toxicities were also seen in animals (non-rodent 63%, rodent 43%; 150 compounds, 221 events; sensitivity only). 13.8% of Phase 1 entrants reach approval (406,038 trial entries, 21,143 compounds; oncology 3.4%). 24153

Reported, not yet replicated

The 86% is not a share more

The "86% concordance" in the same umbrella review is a pooled relative risk of 0.86 (95% CI 0.80 to 0.92): the proportion of positive animal studies divided by the proportion of positive clinical studies, for 62 therapies with at least 5 animal studies. By area it ranges from 0.99 (neurology, 23 therapies) to 0.66 (cancer, 3) and 0.60 (mental health, 7). It is not a share of therapies and not on the same denominator as the 5%. 1

Reported, not yet replicated

The "over 90% fail" statement more

FDA documents say over 90% of drugs that appear safe in animals fail, without a denominator or definition of "appear safe". The one footnoted source opened (Mak 2014) does not contain the 90% and says translation to clinical cancer trials is "less than 8%". Phase 1 to approval is 10.4% in an earlier study and 13.8% in Wong; those are human attrition, not animal-positive attrition. Phase 2 success differs by method: 32.4% in the earlier study versus 58.3% in Wong, which Wong attributes to imputing missing trials. No opened primary source measures animal-positive to approval or the fraction of failures animals caused. 673

Reported, not yet replicated

Documented cases of animals misleading more

Corticosteroids in head injury looked beneficial in animals (pooled odds ratio 0.58) and gave no benefit in trials; tirilazad cut infarct volume 29% in animals and worsened human outcomes (six interventions chosen for clear human results). In a standard ALS mouse, at least 50 publications reported lifespan extension and a controlled retest of eight compounds found no survival benefit for any. TGN1412 (2006): all six healthy volunteers had a systemic inflammatory response; CD28 is absent from effector memory T cells in the species used for safety testing, which the authors say almost certainly explains it. A cancer-biology replication project replicated 46% of 112 effects, with positive effects 85% smaller than originally reported. 89101112

Reported, not yet replicated

Human versus mouse genomic response more

Among genes changed in humans after burns, trauma or endotoxemia, mouse orthologue responses showed R2 0.0 to 0.1 (Seok 2013). A reanalysis restricted to genes changed in both species found Spearman 0.43 to 0.68 with 77 to 93% moving in the same direction (Takao 2015). Both are correct; they answer different questions. Critics note the original used one inbred strain, one sex and unfractionated blood. 131415

Contested

How the first human dose is set more

FDA guidance sets the starting dose by finding the no-observed-adverse-effect level in the most appropriate species, converting by body surface area to a human equivalent dose and dividing by a safety factor, normally 10. The conversion is a size correction, not an evolutionary one. 16

Reported, not yet replicated

Still open here

Not yet written: Plain answer, Why it matters, Equation, Real numbers.

Linked from

Sources

  1. Ineichen BV, Furrer E, Grüninger SL, et al.. Analysis of animal-to-human translation shows that only 5% of animal-tested therapeutic interventions obtain regulatory approval for human applications. PLoS Biol (2024). paper; opened 2026-11-17 all records citing this
  2. Hackam DG, Redelmeier DA. Translation of research evidence from animals to humans. JAMA (2006). paper; opened 2026-11-17 all records citing this
  3. Wong, Chi Heem, Siah, Kien Wei, Lo, Andrew W. Estimation of clinical trial success rates and related parameters. Biostatistics (2019). paper; opened 2026-11-17 all records citing this
  4. Contopoulos-Ioannidis DG, Ntzani E, Ioannidis JP. Translation of highly promising basic science research into clinical applications. Am J Med (2003). paper; opened 2026-11-17 all records citing this
  5. Olson H, Betton G, Robinson D, et al.. Concordance of the toxicity of pharmaceuticals in humans and in animals. Regul Toxicol Pharmacol (2000). paper; opened 2026-11-17 all records citing this
  6. FDA year-one progress document on reducing animal testing (2026). U.S. Food and Drug Administration (2026). regulatory, not peer reviewed; opened 2026-11-17 all records citing this
  7. Mak IW, Evaniew N, Ghert M. Lost in translation: animal models and clinical trials in cancer treatment. Am J Transl Res (2014). review; opened 2026-11-17 all records citing this
  8. Perel P, Roberts I, Sena E, et al.. Comparison of treatment effects between animal experiments and clinical trials: systematic review. BMJ (2007). paper; opened 2026-11-17 all records citing this
  9. Scott S, Kranz JE, Cole J, et al.. Design, power, and interpretation of studies in the standard murine model of ALS. Amyotroph Lateral Scler (2008). paper; opened 2026-11-17 all records citing this
  10. Suntharalingam G, et al.. Cytokine storm in a phase 1 trial of the anti-CD28 monoclonal antibody TGN1412. N Engl J Med 355:1018-1028 (2006). paper; opened 2026-11-17 all records citing this
  11. Eastwood D, Findlay L, Poole S, et al.. Monoclonal antibody TGN1412 trial failure explained by species differences in CD28 expression on CD4+ effector memory T-cells. Br J Pharmacol 161:512-526 (2010). paper; opened 2026-11-17 all records citing this
  12. Errington TM, Mathur M, Soderberg CK, et al.. Investigating the replicability of preclinical cancer biology. Elife (2021). paper; opened 2026-11-17 all records citing this
  13. Seok J, Warren HS, Cuenca AG, et al.. Genomic responses in mouse models poorly mimic human inflammatory diseases. Proc Natl Acad Sci U S A (2013). paper; opened 2026-11-17 all records citing this
  14. Takao K, Miyakawa T. Genomic responses in mouse models greatly mimic human inflammatory diseases. Proc Natl Acad Sci U S A (2015). paper; opened 2026-11-17 all records citing this
  15. Osuchowski MF, Remick DG, Lederer JA, et al.. Abandon the mouse research ship? Not just yet!. Shock (2014). review; opened 2026-11-17 all records citing this
  16. US FDA. Estimating the Maximum Safe Starting Dose in Initial Clinical Trials for Therapeutics in Adult Healthy Volunteers. US FDA (2005). regulatory, not peer reviewed; opened 2026-11-17 all records citing this
Where it sits on the map
History of this record (1)

This record has been changed 1 time, on the dates below. The note kept against each change is written for the people who maintain the atlas, in their own shorthand, so it is not shown here; it is in the record itself.

Status
partly written
Last updated
2026-11-17
Record
content/concepts/animal-to-human-translation.json