Clinical operations
Oncology Trial Imaging Reconciliation Before Database Lock
Trace oncology trial scans from acquisition to central review and learn how to reconcile imaging assessments before database lock.
By Lampgen Clinical Research

3 min read · estimated
What the source reported
In an oncology trial that uses blinded independent central review, a scan performed on time can still leave an endpoint assessment unresolved. Site records, the imaging vendor’s records, and the sponsor’s data each show a different step of the process. Reconciling them during study conduct gives teams time to recover usable assessments ahead of a data cut.
Consider a fictional Week 8 CT. The site documents the assessment within the protocol window. The vendor receives the images under an unscheduled visit label, and the sponsor’s data show no central result for Week 8. The first question is where the records diverged. That determines who can investigate and how the issue can be resolved.
FDA’s 2018 imaging guidance recommends that imaging charters address image transfer, receipt tracking, technical adequacy, and how to address missing images when central interpretation is used. It also discusses the handoff of interpretation data to the sponsor. FDA’s E6(R3) Good Clinical Practice guidance recommends traceable electronic transfers and prespecified procedures to confirm and document how datasets are finalized for analysis. [1, 2]
Assessment timing matters for progression-free survival, an endpoint that generally measures time until progression or death. The protocol and statistical analysis plan determine how an unresolved assessment is handled. The operational priority is to document what happened and recover usable data where the study’s procedures permit. [3]
What this means for trial teams
1. Start with the expected assessment list. Compare the visits expected under the protocol with the site’s acquisition dates, vendor receipt and technical review, central read status, and results received by the sponsor. Use study-approved identifiers and visit codes. A reconciliation view should distinguish an assessment that never occurred, images awaiting acceptance, and a read that exists under a different visit. Each needs a different response.
2. Assign the next action at the point of failure. The site and CRA can trace the acquisition date and upload details within their authorized access. The vendor can check whether the submitted files passed technical review and how the visit was assigned. Data management can confirm whether the result appears in the sponsor transfer under the intended visit. The study lead owns coordination and can bring potential endpoint impact to the appropriate decision-makers. Follow the imaging charter and prespecified data procedures for corrections and preserve the study’s blinding controls.
3. Verify the correction and look for recurrence. Give each open exception an owner, due date, and clear evidence of closure. In the fictional Week 8 case, closure means confirming the result is associated with the expected visit in the authorized sponsor transfer. If other sites use the same upload workflow, compare their exception patterns. A recurring visit-label mismatch is a reason to review the instructions or the data mapping with the vendor. Check the next scheduled transfer to see whether the change held.
4. Protect the participant’s time. Verify whether the existing scan can be retrieved and used before the authorized study and clinical teams consider any repeat imaging under the protocol. Travel, time away from work, and caregiving can make another visit difficult. A recoverable transfer problem is an opportunity to preserve data from the original visit and avoid imposing extra work on the participant.
This article accompanies Issue 01 of The Oncology Trial Brief from Lampgen Clinical Research. The practical test for the next study review is to identify who can confirm that a completed scan produced the expected result in the study dataset and who acts when that evidence is missing.
Sources
- FDA • Clinical Trial Imaging Endpoint Process Standards, April 2018
- FDA • E6(R3) Good Clinical Practice, September 2025, sections 4.2.5–4.2.6
- FDA • Clinical Trial Endpoints for the Approval of Cancer Drugs and Biologics, December 2018
- Lampgen • A worked SIGNAL™ monitoring case
- Lampgen • Clinical trial participation burden and patient access