RQA GLP Committee
As you say, it is the Study Directors responsibility to have confidence in the test item (and they should document their acceptance of the test item in the study data). QA will review this data at some point during the study – maybe at dose administration, maybe at report audit, but unless QA are working in the facility generating the C of A or re-test data then they will not be directly involved with approving such information.
If the re-test data is not being conducted in a GLP facility then QA would not be necessary.
QA should check that there is a process to ensure that someone (the SD) conducts the check, but they do not do it.