New approach to clinical trials and drug registrationBMJ 2001; 323 doi: https://doi.org/10.1136/bmj.323.7308.341 (Published 11 August 2001) Cite this as: BMJ 2001;323:341
Author's suggestions for drug approval are questionable
- Silvio Garattini, director,
- Vittorio Bertele, head of regulatory policies laboratory
- “Mario Negri” Institute for Pharmacological Research, 20157 Milan, Italy
- Clinical Research Consultants, Basle 4058, Switzerland
EDITOR—Jones's proposal for a new approach to drug registration1 echoes industry views2 but hardly takes account of patients' interests, which are to have effective, safe, and cheap medicines.3 Experience indicates that phase I and II studies cannot replace phase III studies.
Advances in pharmacotherapy recently have been based on large randomised clinical trials in cardiovascular medicine, adopting a pragmatic approach with hard outcome measures. Much less has been achieved in oncology, where new substances have been tested in small, phase II, often uncontrolled studies, addressing surrogate or subjective end points. With a few hundred patients, as in phase I and II studies, assessing the true incidence of adverse reactions is impossible and establishing any meaningful ratio of benefit to risk is difficult.
Regulators must seek a balance between denying patients allegedly useful drugs and giving them immature products. Although their decisions mainly rely on randomised clinical trials, sometimes they grant conditional approvals …