> ## Content Index
> Fetch the complete content index at: https://www.fdaweb.com/llms.txt
> Use this file to discover other available public pages before exploring further.

# Bad Idea to Eliminate Phase 2 or Phase 3 Trials: Academics
- URL: https://www.fdaweb.com/bad-idea-to-eliminate-phase-2-or-phase-3-trials-academics/
- Published: 2017-02-02T12:00:00.000Z
- Updated: 2026-09-14T22:04:24.000Z
- Author: David McFarland
- Tags: Drugs, #legacy-id-D5137895

It would be unwise to speed up drug development and approvals by reducing or eliminating Phase 2 or Phase 3 trials, according to an [opinion piece](http://www.the-scientist.com/?articles.view/articleNo/48280/title/Opinion--Improving-FDA-Evaluations-Without-Jeopardizing-Safety-and-Efficacy/&ref=fdaweb.com) in *The Scientist*. Written by two academics(\*), the article cautions that such an approach would not test the drugs in a sufficient number of patients to identify health risks. Speeding development and approvals is a top priority for FDA under the Trump Administration ([see earlier story](https://www.fdaweb.com/less-fda-for-lower-prices-trump-to-pharma-ceos/)).

“The FDA could explore using [adaptive licensing](https://www.ncbi.nlm.nih.gov/pubmed/22336591?access%5Fnum=22336591&link%5Ftype=MED&dopt=Abstract&ref=fdaweb.com), whereby a drug is provisionally approved based on Phase 1 and 2 data for a defined short period of time,” the article says. “During the provisional approval period, efficacy data are collected directly from clinical use, accruing [real-world data](http://clinicalresearcher.acrpnet.org/content/30/1/34.article-info?ref=fdaweb.com) capable of overcoming the generalizability limitations of traditional trials. After a year or two, the drug is either fully approved or loses its marketing license.”

Instead of eliminating late-stage trials, “pragmatic and adaptive trial designs could help reduce numbers of research subjects needed to test a drug and increase the likelihood trial data are generalizable to the patient population that will use a drug following FDA approval,” the authors write. “Currently, many trials are conducted on highly specialized patient populations — for example, healthier and younger than typical patients — causing concerns about the quality of our medical evidence at the time a new drug is introduced to market.

Next, the agency could explore [reducing the use of animal models](http://www.the-scientist.com/?articles.view/articleNo/47996/title/Opinion--Reuse-and-Reduce/&ref=fdaweb.com) in preclinical research. “Currently, the FDA requires extensive and sometimes expensive animal trials before entering human studies,” the article says. “While such experiments often detect unwanted side effects, many animal models of diseases are poorly translated into successful human trials. As [human organoid technology](http://jcb.rupress.org/content/early/2016/12/27/jcb.201610056?ref=fdaweb.com) continues to be developed and improved, these mini human organs may prove to be a better and cheaper model system to test the effects of drugs instead of using expensive and often inaccurate animal models.”

Additionally, the authors recommend that academic institutions and pharmaceutical companies should be required to publish the results of all clinical trials within one year of a trial’s primary completion date and to make their patient-level data available to all researchers and physicians. Currently, only 57% of trials for new drugs are registered and only 65% have [publicly disclosed results](http://bmjopen.bmj.com/content/5/11/e009758.full?ref=fdaweb.com). Fostering more open science is likely to save money and spur innovation, as researchers learn from the lessons of others and avoid duplicating costly trials.”

  
*\* *John D. Loike* is a faculty member at Columbia University College of Physicians and Surgeons. *Jennifer Miller* is a faculty member at New York University School of Medicine and the President of Bioethics International*