Failing to obtain informed consent prior to randomization
Subjects were added to IVRS prior to consent and could not be removed from the system when parents would not consent
Enrolling subjects who were not eligible
Randomizing three subjects to the wrong study arm
Administering investigational product that had undergone a 30 minute temperature excursion
Failing to perform study required follow up phone calls
Three subjects received prohibited concomitant study medications
EDC audit trails show that procedures originally entered as not done, and not documented in source records were changed to having been done due to a reason of ‘entry error’.
Consent documentation errors, including:
Signing consents prior to the subject’s parents signed
Failing to retain original documents
Changing the time of the consent to appear that consent occurred prior to dosing
FDA did not post any warning letters to Sponsors/Monitors/CROs, IRBs or Clinical Investigators this week. From the archives, here is the Clinical Inspection Summary for Alunbrig (brigantinib) submitted by Ariad. Four sites, plus the Sponsor and Contract Research Organization (CRO) responsible for the independent reads of the trial images were inspected. Site inspections ranged over 4 to 9 days, the Sponsor inspection took place over 4 days. One site and the CRO were issued FDA Form 483s.
Dr. Kim was cited for:
- Failing to follow the delegation of authority log. Specifically, the study coordinator assessed adverse events
- Failing to document hand and arm pain as adverse events in one subject
- Failing to permit FDA to copy reports. Specifically, the audit trail of the electronic medical record could not be printed
The unnamed CRO was cited for using a computer system prior to validation, specifically:
- Implementing the software (Mint lesion) at a time when the vendor had disclosed that it was not 21 CFR 11 compliant
- Validating the software without confirming that it complied with charter requirements for lesion measurement. (Note: images for 126 subjects were required to be reread due to errors by the software.)
- The CRO did not comply with charter requirements for variability testing
- Inadequate audit trails that did not capture information required by the charter for reasons items were reread.
- Quality assurance checks for incoming images from clinical sites were not conducted in the timeframes required by the charter
FDA notes that due to the issues noted at the CRO in the inspection, the validity of the data from the CRO (the trial's primary endpoint) could not be verified