Achieving Critical Database Lock Milestones in a Global Oncology Study 

By Marloes Theusink, MSc (Chemical Engineering), Senior Project Director, Global Project Leadership at Allucent

Study Background

Allucent was selected to provide full-service support for a Phase IIb randomized clinical trial in patients with first line metastatic head and neck squamous cell carcinoma (HNSCC) for a leading global biopharma company.

The global study was conducted across 31 sites in multiple countries and regions and enrolled 171 patients. Key study activities included management of complex oncology efficacy and imaging endpoints, ongoing source data verification (SDV), query management, coding activities, and database lock coordination.

A critical operational objective was to keep the study on track for two planned database lock milestones: the interim primary analysis database lock in May 2024 to support planned publication of interim efficacy findings, and the final database lock in December 2025. With last patient last visit (LPLV) expected in November, the final lock required close coordination to complete all database lock activities within five weeks of LPLV and avoid delays associated with year-end holidays and reduced site availability.

Scope of Final Database Lock Activities

• Approximately 28,000 data fields subject to source data verification (SDV)
• 8,366 coded terms requiring management and reconciliation
• More than 900 imaging assessments requiring submission and oversight

Operational Challenges in Maintaining Database Lock Readiness

Maintaining high-quality data throughout the study was particularly critical because a central imaging review process was not included in the Sponsor’s study design. Delays in site data entry, inconsistent source documentation, and poor source data quality had the potential to create backlogs and impact overall database integrity and robustness.

Operational complexity increased further due to site-specific logistics affecting tumor evaluation turnaround times for scans. Ensuring consistent application of tumor response evaluation criteria, including RECIST 1.1 and iRECIST, across all participating sites added another layer of complexity.

Additionally, the protocol deviation (PD) reporting and review process in place at study initiation relied partially on manual workflows, requiring CRAs to review multiple sources while balancing broader site management responsibilities. This created inefficiencies and increased the risk of backlog accumulation.

The timing of LPLV also introduced an additional operational challenge. Because LPLV occurred in November, the study team faced compressed timelines to complete database lock activities within five weeks to avoid delays associated with year-end holidays and reduced site availability.

Proactive Strategies to Strengthen Data Quality and Database Readiness

Experienced CTLs and CRAs with expertise in tumor-specific and immune-related response evaluation criteria, including RECIST 1.1 and iRECIST, were assigned to the study to support consistent data quality and efficacy assessment oversight across sites.

To strengthen source documentation quality, CRAs provided tumor assessment worksheets to sites that did not have standardized tools in place. Sites received retraining when data quality issues were identified, and ad hoc quality assessment visits were conducted to reinforce source data accuracy and protocol compliance.

To promote timely site data entry and scan submissions, the clinical team implemented a tracking tool that calculated scan due dates based on each patient’s first treatment cycle. This enabled CRAs to proactively issue timely reminders to sites throughout both the treatment and follow-up periods.

A risk-based monitoring strategy was implemented through centralized monitoring activities and study-specific Key Risk Indicators (KRIs), enabling the team to identify higher-risk sites and apply targeted oversight. Study data metrics and site performance were reviewed weekly to rapidly identify issues and implement corrective action plans when needed.

Ongoing site engagement was maintained through global communications, newsletters, and ad hoc site calls focused on database lock expectations and study timelines.

In parallel, the protocol deviation process was transitioned to a fully automated workflow within Veeva Vault, including Sponsor review and reporting activities. This streamlined protocol deviation management process, reduced manual burden, and eliminated backlog risks. Third-party data reconciliation activities were also conducted on an ongoing basis to minimize late-stage queries and support a smooth database lock process.

Results: Delivering High-Quality Data and Accelerated Database Lock Timelines

These efforts resulted in a robust and reliable dataset, with the final database lock achieved on target just five weeks after LPLV. Successful execution of the interim primary analysis database lock supported publication of interim findings describing objective response rate (ORR) in the PD-L1 low population. Data entry backlogs were minimized, and the overall source data verification (SDV) rate was maintained at 90–95% throughout the study, reaching 100% at the final database lock.

The focus on timely data entry, scan submission, and query resolution supported high-quality efficacy data and overall database readiness. Consistent application of tumor response evaluation criteria, combined with ongoing site training and oversight, further strengthened data accuracy across participating sites.

In parallel, the upgraded protocol deviation management process eliminated manual inefficiencies, while ongoing third-party data reconciliation minimized late-stage queries and supported a smooth and timely database lock process.

Overall, the study successfully met critical operational milestones while supporting the Sponsor’s development objectives in a complex global oncology trial.  Achieving database lock within five weeks of LPLV across 31 global sites demonstrates the study team’s ability to effectively manage complex operational requirements, multiple data streams, and demanding timelines while maintaining data quality and integrity.

By providing high-quality data within critical decision-making timelines, the study team supported the Sponsor in advancing their publication strategy, communicating emerging results to key stakeholders, and making strategic prioritization decisions.