Partner Investigation Adequacy Review Checklist

Document type	Checklist
Topic area	CDMO Governance
Level	Advanced
Updated	September 2026

This is a ready-to-use checklist for reviewing a contract manufacturer’s deviation investigation before the sponsor accepts its root cause and corrective action, so a partner’s write-up is graded against a defined standard rather than accepted because it reads confidently. Complete this against the raw investigation, not the forwarded summary. Replace every <<FILL: ...>> placeholder with your own specifics. A filled specimen follows.

Document control header

Field	Entry
Document title	Investigation Adequacy Review for `<<FILL: deviation ID>>`
Document number	`<<FILL: e.g. IAR-2026-011>>`
CDMO / site	`<<FILL>>`
Deviation reference	`<<FILL: partner's own deviation number>>`
Reviewer	`<<FILL: e.g. Supply Chain Quality lead>>`
Review date	`<<FILL>>`

1. Purpose

This checklist confirms, before a sponsor accepts a contract manufacturer’s investigation as closed or acts on its stated root cause, that the causal chain actually terminates on something correctable, that relevant trend data was checked, and that the proposed corrective action addresses the control gap rather than the operator alone. It is a gate on acceptance, not a courtesy review after the fact.

2. Scope

Dimension	Decision
Deviation under review	`<<FILL>>`
Investigation document(s) reviewed	`<<FILL: reference the raw investigation, not only the forwarded summary>>`
Out of scope	`<<FILL: e.g. the sponsor's own cross-network impact assessment, tracked separately>>`

3. Responsibilities

Role	Responsibility
Supply chain quality / CDMO quality lead	Owns this review, decides whether to accept, request rework, or escalate
Person-in-plant or resident quality representative	Provides first-hand floor context where available
Regulatory affairs	Confirms whether the deviation, once its root cause is accepted, carries any reportability implication

4. Adequacy review items

#	Item	Pass / Fail / NA	Evidence
1	The causal chain terminates on something correctable (an equipment, method, or systemic gap), not a first plausible-sounding explanation	`<<FILL>>`	`<<FILL>>`
2	The investigation states why the control that should have caught the failure did not	`<<FILL>>`	`<<FILL>>`
3	Trend data was checked: whether this exact failure mode has fired before, at this site, on this line	`<<FILL>>`	`<<FILL>>`
4	If this failure mode has fired before, the investigation explains why the prior CAPA did not prevent a repeat	`<<FILL>>`	`<<FILL>>`
5	The sponsor reviewer read the raw investigation record, not only the forwarded summary	`<<FILL>>`	`<<FILL>>`
6	Where a finding was ambiguous, the investigator or operator was consulted directly rather than the write-up accepted as the full record	`<<FILL>>`	`<<FILL>>`
7	The proposed corrective action targets the control gap itself, not only retraining or reminding the operator	`<<FILL>>`	`<<FILL>>`
8	An effectiveness metric, observation window, and accountable owner are defined for the CAPA before approval	`<<FILL>>`	`<<FILL>>`
9	The investigation states, or the reviewer has separately assessed, whether the same process step runs at another site in the sponsor’s network	`<<FILL>>`	`<<FILL>>`

Any item scored Fail blocks acceptance of the investigation as closed. State the required rework or the escalation ladder rung invoked instead.

5. Disposition

Field	Entry
Overall disposition	`<<FILL: accepted / rework requested / escalated>>`
If rework requested, what was asked for	`<<FILL>>`
If escalated, ladder rung invoked	`<<FILL: reference the escalation ladder in the deviations and change control article>>`
Follow-up owner and due date	`<<FILL>>`

6. Approval

Role	Name	Signature	Date
Reviewer	`<<FILL>>`
Supply Chain Quality lead (if different)	`<<FILL>>`

7. References

FDA guidance, “Contract Manufacturing Arrangements for Drugs: Quality Agreements” (2016). ICH Q10, Pharmaceutical Quality System, Section 2.7, Management of Outsourced Activities and Purchased Materials. ICH Q9(R1), Quality Risk Management.

Confirm the current version of each reference before issue.

8. Revision history

Version	Date	Author	Summary of change
`<<FILL: 1.0>>`	`<<FILL>>`	`<<FILL>>`	Initial issue.

Filled specimen

Review of CMO Beta’s investigation into a fill-line underweight deviation on Acme Therapeutics’ ATX-330 drug product, three months after a prior underweight deviation on the same line.

#	Item	Pass / Fail	Evidence
1	Causal chain terminates on something correctable	Fail	Investigation stopped at “operator error, retrained”; no explanation of why the in-process weight check did not catch the trend before four units were affected
3	Trend data checked	Fail	No reference to the prior underweight event three months earlier on the same line; reviewer’s own trend pull found it in under ten minutes
5	Raw investigation reviewed	Pass	Reviewer requested and read the full investigation, not the one-paragraph summary CMO Beta initially forwarded
7	Corrective action targets the control gap	Fail	Only action proposed was operator retraining; no change to the in-process weight check frequency or alarm threshold that should have caught the drift

Disposition: rework requested. CMO Beta was asked to reopen the investigation, address the repeat-event trend explicitly, and propose a corrective action tied to the in-process check itself, not the operator alone, per items 1, 3, and 7 above. Follow-up owner: Supply Chain Quality lead, due in 10 business days. The pattern, two underweight events on the same line in a quarter, was also raised at the next joint quality review per the escalation ladder, since a single-incident conversation had already occurred once for the first event without resolving the underlying gap.

Common inspection findings this checklist prevents

A partner investigation accepted at face value with no evidence the sponsor reviewed the raw record.
A repeat failure mode with no trend check on file, so the same root cause gets rediscovered, and re-explained, event after event.
A corrective action that addresses the operator and leaves the control gap that should have caught the failure untouched.
No effectiveness metric, window, or owner defined at the time the CAPA is approved.

How to adapt this checklist

Set the deviation reference and CDMO site in the header.
Score every item against the raw investigation record, not a summary; leave no item blank.
Any Fail blocks acceptance; state the specific rework requested, not a general request to “improve” the investigation.
Log the disposition and follow-up owner in section 5, and carry a repeat Fail on the same failure mode into the escalation ladder.
Confirm every reference in section 7 against the current published version before issue.