ZEPHYR-101 WEEKLY STUDY TEAM MEETING NOTES (SYNTHETIC)
Week of 28 September 2026. Fictional study and fictional product ZX-204. Not real data.

Attendees (roles only): study lead, lead statistical programmer, biostatistician, data manager, safety scientist, clinical operations lead, regulatory submissions manager.

1. TIMELINE
- Database lock has been brought forward from 30 October to 23 October at the sponsor's request. Not everyone agreed it is achievable.
- Study lead asked whether the second dry run can still happen on 22 October. Nobody committed to this.
- The review window before lock is currently five working days. Clinical operations assumes it stays five. Programming assumed it would be reduced to three if needed.

2. DATA
- Central lab: final transfer is expected 9 October but the vendor has hinted at a slip of up to four days. Data manager will ask for a written date.
- ECG vendor: final transfer date not yet confirmed in writing.
- Data manager reported 11 lab records that do not reconcile between the vendor file and the clinical database. Cause unknown.
- Imaging reader adjudication is still running; the date it will finish was described as "soon".

3. PROGRAMMING
- Lead statistical programmer reported that independent QC of the adverse event and subject-level analysis datasets is about two days behind plan.
- Two analysis datasets use different baseline reference dates. Biostatistician believes the subject-level dataset is correct; programming believes the lab dataset follows the specification. This is unresolved.
- Treatment-emergent flag logic differs between two datasets; fix is in progress.
- Reviewer guide drafts have not started.

4. SAFETY
- Safety scientist asked for the adverse event overview table before the second dry run. This was not previously on the plan.
- A decision is needed on whether late adverse event data after the cut will be included in lock outputs.

5. DECISIONS MADE
- Weekly findings review will move to Tuesdays.
- Data manager owns the written confirmation of transfer dates.

6. ACTIONS
- Data manager: obtain written final transfer dates from the lab and ECG vendors. Due: 7 October.
- Biostatistician and lead programmer: agree the baseline reference date and document it. Due: not set.
- Lead programmer: confirm whether independent QC can recover the two days. Due: 8 October.
- Study lead: decide whether to protect the five-day review window. Due: next meeting.
- Regulatory submissions manager: start the reviewer guide outline. Due: 12 October.
- Safety scientist: confirm the adverse event table specification. Due: 9 October.

7. OPEN QUESTIONS
- Who has authority to approve a shorter review window?
- Is the earlier lock date confirmed in writing by the sponsor?
- Will the late change request (see separate memo) be included before lock?