CIOMS Working Group X on "Considerations for applying good meta-analysis practices to clinical safety data within the biopharmaceutical regulatory process"
Meta-analyses have been increasingly used in the scientific evaluation of efficacy and safety in benefit-risk assessments, as well as for pricing and the reimbursement of biopharmaceutical products. Over the last few years, the outcome of meta-analyses has been both debated and used as evidence within the regulatory process for human medicinal products, and their interpretation has sometimes contributed to different regulatory actions for the same product.
Working Group X was established, based on suggestions from several parties, and consists of senior scientists from drug regulatory agencies, the biopharmaceutical industry and academia. The objective is to develop a consensus on scientific and methodological criteria that represents good practices when applied to meta-analyses of clinical data within the regulatory process. The aim is that these criteria will then be used by both industry and regulators. During its first meetings the Working Group decided to focus mainly on meta-analyses of safety data and the title of the Working Group was revised to reflect this.
The members of the Working Group have identified that with increasing amounts of data available in digitized form, publications of meta-analyses of clinical data will likely increase. Apart from methodological caveats, the access to relevant unpublished studies and data submitted only to regulatory agencies or remaining in the files of biopharmaceutical companies, availability of sources of data, the possibility to interact with the researchers involved in the original studies and the resources to re-analyze available data will need to be considered. The ambition is to try to develop a recommendation of a harmonized view on these issues.
In addition, the combining of available information from both randomized clinical trials and epidemiological studies to generate an integrated result is controversial. Criteria as to how this should be done, particularly for safety data, needs to be elaborated and general principles should be established.
It is envisioned that the Working Group will have a series of meetings over a period of three years, aiming for a publication in 2015.