T-02: Using Innovative Automation to Author Development Safety Update Reports and Enhance Cost-Effectiveness
Poster Presenter
Nipa Parikh
Executive Director, Global Head of PV Strategy and Operations
Otsuka Pharmaceutical Development and Commercialization Inc. United States
Objectives
Develop a Development Safety Update Report (DSUR) authoring tool on a computerized collaboration platform to auto-populate a text-based business template, to reduce writing efforts without compromising quality or compliance, thereby saving time and money.
Method
Using a data analytics model, section requirements were mapped to relevant text-based templates. With the use of expert and text rules as well as natural language processing and generation, data and text were merged without manual intervention.
Results
Of the 20 major and 11 minor DSUR sections, the team was able to fully automate nine sections (including: Title Page, Worldwide Marketing Approval Status, Status of Clinical Trials Ongoing and Completed during the Reporting Period, Cumulative Subject Exposure in the Development Programme, Data in Line Listings and Summary Tabulations and sub-sections Reference Information, Interval Line Listings of Serious Adverse Reactions during the Reporting Period, Cumulative Summary Tabulations of Serious Adverse Events, and Region-Specific Information) and partially automate 12 sections (including: Executive Summary, Introduction, Patient Exposure from Marketing Experience, Completed Clinical Trials, Ongoing Clinical Trials, Long-term Follow up, Other Therapeutic Use of Investigational Drug, New Safety Data Related to Combination Therapies, Other Clinical Trial/Safety Information, Safety Findings from Marketing Experience, Other DSURs, and Summary of Important Risks).
In 2017, seven DSURs were developed, five using the automation tool and two using complete manual intervention. Total time saved was measured by comparing manual versus computer-facilitated authoring times. This resulted in a total time-saving of 26 h per DSUR (without automation: 305 h; with automation: 279 h), with an annual time-saving of 182 h. The total time savings was calculated based on the number of hours it took to complete authoring of the relevant sections. This was determined by conducting a time study for two DSURs and then taking the average of the total hours per section. This translated to an annual cost-savings of $7000 per report, or $35000 for the year of 2017. In 2018, seven DSURs were developed, of which all used the automation tool, thus resulting in a cost savings of $49,000.
There was no decrease in the quality of the automated versus the manually prepared DSURs. In addition, compliance was 100%.
Conclusion
The DSUR automation tool enabled consistency with the production process by following the same template, data sources, and steps needed to generate a report. Furthermore, any process which is automated significantly reduces the chance for human error, while minimizing the impact of the different authoring styles which contributors may use. Therefore, this allows for a more standardized and harmonized approach. Out of the ten sections which could not be currently automated, five sections required analysis of safety data.
Thus, the time saved by automating or partially automating approximately 2/3 of the report allowed for more time to focus on safety analysis during authoring. In conclusion, the tool enhanced the potential of the organization to generate large volumes of quality reports with limited resources in a timely manner and improved the overall quality and consistency of DSURs. It transformed a resource heavy and time-consuming process to a more streamlined and cost-effective process.