The data exists in the clinical record for each patient and can in theory be retrieved, the question might be at what cost. If the N count doesn’t have a large number of placeholders it would be more doable, especially if you were only missing administration dates for a few patients. The trial site typically submits the data required through a reporting portal to the CRO who in turn is supposed to conduct trial site audits as well as an auditing of the data submitted for proper formatting, missing fields etc. before database lock. Amarex clearly has a lot to answer for but make no mistake, just because the Amarex database might have gone through a blender or lack some ingredients doesn’t mean the clinical data is not discoverable…it is.