Drug Evaluation Committee 2013-39 Anonymization of pharmacogenomics samples
Related classification: Other
Initial release date: November 2013
Revised release date: March 2021
Question
Pharmacogenomics (PGx) content is increasingly included in global clinical trials. We have a question regarding the anonymization of samples for PGx in the "Considerations for Pharmacogenomics in Clinical Trials of Pharmaceuticals (April 2018)", especially regarding the storage of the corresponding table at the site when further coding is not performed by the sponsor (and the genetic analysis organization). I have a question regarding the storage of the correspondence table at the site, especially when further coding is not performed by the sponsor (and the genetic analysis organization). The "About Clinical Trials Using Genome Pharmacology (September 30, 2008, Pharmaceutical and Food Safety Bureau, Ministry of Health, Labour and Welfare, Notification No. 0930007)" does not describe specific operations, so we refer to the above document of the Pharmaceutical Manufacturers Association of Japan (PMAJ).
The above document states that either single code or double code is used for linkable anonymization, and that single code is the "standard for clinical trials. The correspondence table between personal identifiers and subject identification codes that is usually used in clinical trials "corresponds to the list of subject identification codes that is a GCP-required document" and therefore, it is assumed that the correspondence table that is usually used in clinical trials is used for single coding and that the requirement for linkable anonymization for PGx samples has been met. Is it appropriate to treat PGx samples in a clinical trial in this way? Or, is it necessary to create a new correspondence table that links a list of subject identification codes to another code within the implementing medical institution? We would appreciate your answer from the viewpoints of handling of personal information including genomic genetic information and blinding.
The person responsible for keeping the correspondence table states that "the correspondence table between personal identification information and subject identification code corresponds to the list of subject identification codes, which is a GCP-required document". Is it correct to say that this correspondence list should be kept by the investigator as in a normal clinical trial? It is stated that "The person who maintains the correspondence table between single code and double code must not be in a position to know the 'list of subject identification codes'", but does this also apply to single-code linkable anonymization? If so, should it be done by a clerk or other person who is not in a position to know, since the investigator or staff of the clinical trial management office are in a position to know?
Opinion of the Pharmaceutical Manufacturers Association of Japan (PMAJ)
When PGx analysis data collected in a clinical trial is used as material for a drug application, the PGx sample should be anonymized in a way that it can be traced back to the source material of the data. The PGx anonymization at the site should be up to single coding, and the double coding and management of the correspondence table between single and double codes should be performed by the sponsor or an external organization entrusted by the sponsor.
We believe that it is acceptable to use the "list of subject identification codes," which is a document required to be kept under GCP, as a list of the correspondence between codes and subjects' personal identification information that should be kept at the site. The person in charge of keeping the corresponding list should be the person in charge of keeping the "list of subject identification codes" in the clinical trial (usually the investigator).
Please refer to "Glossary of Terms in Genomic Pharmacology" (January 9, 2008, Pharmaceutical and Food Safety Bureau, Ministry of Health, Labour and Welfare, Pharmaceutical and Food Safety Bureau, Director-General's Notice No. 0109013 / Pharmaceutical and Food Safety Bureau, Director-General's Notice No. 0109002) (ICH E15) for single code and double code.
Reason for revision of opinion
In accordance with the revision (April 2018) of the Pharmaceutical Manufacturers Association of Japan (PMAJ) recommendation document "Items to be considered when conducting pharmacogenomics in clinical trials of pharmaceutical products," minor changes have been made to the questions.