Recently, the Veramed team addressed the challenge of a delivery with tight timelines using automation. In this case, the team focused on automating the TFL and increasing the efficiency of Lab outputs. In clinical trials, a blind data review, also known as a dry-run, allows the statistical tables, figures and listings (TFL) to be evaluated prior to database-lock. While the primary purpose is to verify the assumptions and analysis methods detailed in the Protocol and Statistical Analysis Plan (SAP), it also provides an important opportunity to assess the trial in terms of e.g.
- Recruitment.
- Population sizes and rates.
- Protocol deviations.
- Potential safety issues.
As a result, the Protocol or SAP may need to be amended, which makes the first blind review a critical milestone.
The Challenge
The creation of all the analysis datasets and TFL required for the first dry-run can often take the statistical programming team a considerable amount of effort. The majority of the outputs for the final CSR are often required. However, at the same time, the trial is often in an early stage where there is sparse amounts of data on a relatively small number of subjects.
The major challenge highlighted was finding out how the team could deliver the TFL for the first dry-run as early as possible and allow meaningful decisions to be made, while at the same time ensuring the level of quality required to make informed decisions.
The Solution
In this study, safety reporting of lab data was a key part of the initial analysis, with many repeats of Hematology, Chemistry, Coagulation and Urinalysis summary and shift-from-baseline tables.
The Veramed automation team worked closely with the Veramed delivery team to provide an integrated solution that produced standalone TFL programs to expedite the validation of the Lab outputs.
This involved creation of CDISC ARM v1.0 standards-based metadata, which was then used to drive the automation of standalone SAS programs for the Lab summary tables. This in turn allowed the Veramed delivery team to accelerate the validation of the outputs following industry-standard, parallel programming practices.
Outcomes
By automating the creation of repeating standard Lab outputs, Veramed’s clinical Statisticians and Programmers were able to focus on the study-specific issues needed to deliver a high-quality dry-run for a dynamic Biotech client. As a result, the team was able to drastically increase efficiency and ensure on-time delivery that was within budget.
Keen to learn more? Discover our automation and technology services below. Attending PHUSE US in Orlando this March? Don’t forget to drop by Booth 5 to learn more about our innovative processes and technology that have helped Veramed clients worldwide.
