Clinical trials in drug development: a minimalistic approach.

PURPOSE OF REVIEW

Drug development in oncology finds itself at the crossroad of unique opportunities and major challenges. The old paradigms should and can be replaced by a system that better matches the right patients to the right compounds and puts much more emphasis on the early stages of drug development.

RECENT FINDINGS

The clinical phases of drug development will no longer be split into phase I, II, and III studies, but rather into 'functional target pharmacology studies', followed by 'proof of concept studies'. The resulting development flow becomes Apollo-capsule shaped.

SUMMARY

Although randomized studies will still be needed for drugs using targets in the tumor environment, or for combinations of agents, drug registration might proceed without these if all of the following criteria are met in early development: availability of preclinical convincing evidence that the drug's target is the functional driver behind the disease phenotype, availability of a predictive biomarker that enables appropriate and actual patient selection in early pharmacology studies, a Response Evaluation Criteria In Solid Tumors (RECIST)-based single agent response rate of at least 50%, and/or a progression at first tumor assessment rate of 15% or less, a duration of absence of progression (stable disease) beyond doubt and considered clinically relevant, and no major safety concern. This set is not yet mature, but may be adapted over time. The concerns related to registering agents on the basis of small datasets can be adequately addressed by obligatory postmarketing hypothesis driven studies.