Forensic Pharmacometrics

Protect yourself and your company’s assets against dataset errors, incomplete analysis plans, and disorganized reports before you submit your package to a regulatory agency. At Simulations Plus, we’re experts in all things related to model-based research and drug development. We have used proactive quality assurance techniques on our own modeling and simulation projects for over 25 years, and we are now offering this service to you. As pharmacometric modeling and simulations become more and more integral to critical path decision making, our forensic review and feedback will help you avoid costly surprises.

Analysis Plans

Data analysis plans for pharmacometric analyses can be missing major content, including descriptions of planned exploratory data analyses, biomarker sampling strategies, designs for the studies contributing data to the modeling effort, planned covariate analyses, and methods used to generate individual exposure measures. Simulations Plus has developed canonical documents that contain standard sections and content suggestions to facilitate adherence to content standards. Forensic assessment of your data analysis plan by Simulations Plus will give you confidence that your plan is complete, accurate, coherent, and ready for regulatory scrutiny.

Datasets

Dataset errors are often subtle and difficult to detect. Forensic assessment of your dataset by Simulations Plus will ensure the dataset was built correctly, confirming that valid and appropriate assumptions were applied, and no critical errors were made in the processing and pooling of study data. In addition, a forensic assessment can determine the adequacy of the data for the analyses. Our experts can perform non-programmatic data checks, such as requirements management, code review, program verification, and comparisons to raw data. Our scientists can also perform programmatic data quality checks to detect potential outliers, to verify record counts and listings against expectations to confirm accuracy, and to generate and review graphical displays that will assist in determining if errors were made in the manipulation of the data.

Pharmacometric Technical Reports

Well-written and organized technical reports can make your submission stand out with the regulators. Simulations Plus has developed canonical report templates to promote inclusion of content recommended in regulatory guidances. Forensic assessment of your submission by Simulations Plus can ensure that your technical reports include these common omissions: influence of study dropouts, description of planned exploratory data analyses, rationale for selecting covariates and for deciding which covariates were evaluated on which parameters, and methods for calculating derived variables.

02. Explore
A picture is worth a thousand words ….

Simulations Plus has developed an extensive graph library with built-in standards and options that allow us to quickly generate pictures of data pooled from multiple studies. The graph library generates:

  • Diagnostics that can be used as part of quality control efforts,
  • Informative guides to direct model building efforts, and
  • Graphical displays of hard-won insights to tell the story of  your analyses to R&D teams.

Pharmacometric modeling often involves the pooling of data from multiple studies, in some programs up to 20 or more studies. Just running a model up against 20 studies, without having the graphical insight into your data is nothing more than a guessing game, and a black box to the extended R&D team. Yet, graphing the data from 20 studies with a time-crunched deadline seems impossible and will cost valuable time.

At Simulations Plus we have developed a suite of processes, requirement standards, templates, macros, and technologies that allow our modelers to be able to spend the time thinking about your data by looking at the exploratory graphical analyses with your extended team. Through this process, we gain critical insights into understanding what the impact on the model will be and what it means to your development program. With this knowledge and insight, our modelers are able to systematically approach model building to efficiently achieve the best predictions possible for your critical decision-making milestones. 

The NONMEM Users Group archive is no longer being maintained on this website.