By Daniel Weiner
June 23, 2008 | Pharmacokinetic/pharmacodynamic (PK/PD) data managers and scientists involved in early drug development are being squeezed by several trends: an accelerating pipeline of candidate drugs emerging from discovery requiring in vivo PK evaluation; increasing demand for more elaborate and intensive PK/PD evaluations at earlier stages of development; and the FDA’s Critical Path Initiative which is calling for more use of PK/PD modeling. As a result, leading companies are meeting these challenges by implementing a clinical pharmacology data repository that provides greater standardization and access to data and improves productivity by leveraging the time of scarce scientific talent.
Clinical Pharmacology Workflow
The production of compliant reports from PK/PD datasets by drug metabolism and pharmacokinetics (DMPK) and clinical pharmacology staff is a significant consumer of effort, time, and cost. Our experience with hundreds of kineticists over the past 15 years indicates that total work effort splits about evenly into three key subprocesses: 1) dataset creation; 2) analysis; and 3) reporting.
More than 20 companies have now invested in clinical pharmacology data repositories–a single, secure storage medium where all clinical pharmacology information is maintained in a consistent format. The repository will typically offer tight integration to PK analytical tools and permits the building of an efficient, electronic means of regulatory-compliant PK/PD analyses and report production (automatic recording of changes, password control, audit trails, etc.).
A clinical pharmacology data repository is typically implemented as the heart of a regulatory-compliant PK workflow system, including data retrieval, data analysis, and reporting. Software called “connectors” link the repository to laboratory information management systems (LIMS) and other PK data sources, automating much of the data retrieval. Many standard PK analyses can be scripted with built-in business and formatting rules.
Reporting within the repository can be automated with software tools that push studies tables, figures, and listings into Microsoft Word. Reporting tools can flag any reported items that are out of sync with the source data and permit rapid “refresh” of reports, if necessary, to bring the report back into sync with the data. The entire system automates many manual reporting tasks, reducing errors and freeing scarce scientific resources to focus on higher value activities such as modeling and simulation.
Compliance
All companies using electronic record systems for FDA submission must comply with 21 CFR Part 11. For companies electing to use electronic record systems the question is not whether, but how, compliance will be achieved. Will it be through manual methods that consume great amounts of scientific time, or will it be through electronic compliance systems?
One aim is to use a repository to provide streamlined QA Audit and QC review procedures in a Part 11 compliant system. The data repository thus assists in compliance. The benefit is a compliant environment in which to analyze and store data and analyses. This replaces manual methods that would be much more time consuming to reach the same result.
Higher Productivity
PK/PD modelers use time-concentration data in modeling and simulation to support development decisions, such as dose optimization or “go/no go”. Although these development decisions do not require "regulatory compliant" data, more organizations are taking the viewpoint that modeling and simulation should be performed to the higher regulatory standard.
The typical drug development company performs dozens or even hundreds of trials per year that generate PK/PD data. PK/PD scientists often must retrieve data from multiple sources and/or locations. A great deal of time is spent rearranging and transforming time-concentration data sets that have been constructed in slightly different ways across various studies, projects, and organizations. Time spent manipulating data sets is time not available for higher scientific analysis.
Better PK analysis accelerates decision-making and improves decision quality. Better traceability means that less time must be spent by scientists in manual regulatory work, and more time can be devoted to understanding the key issues in advancing a drug toward registration.
Unplugging the Biostat Bottleneck
It is a reality in many companies that scientists in DMPK and clinical pharmacology must rely on biostatistics to provide merged, analysis-ready datasets for PK/PD analysis and reporting. Often data requests take days or weeks, and delays can result because biostatistics staff must cope with time demands from multiple programs.
A fully functioning clinical pharmacology data repository frees the organization from the “biostat bottleneck.” Datasets are automatically merged and more or less instantly available. This makes possible more timely interim and final analyses, and more informed decision-making.
Clinical pharmacology data repositories are meeting the need for high-quality, regulatory-compliant production of standardized PK analyses and reports. It is supporting modeling and simulation by making PK/PD data more readily available so that modeling inputs can be produced earlier, with greater efficiency, and without the workflow problems inherent in the “biotstat bottleneck.”
Implementing a repository creates a number of concrete benefits, including standardization of analysis rules, definitions, and formats. Standardization leads to automation, which can further increase productivity of DMPK staff. Compliance is achieved by careful definition and management of workflows and a client-customized validation process. A clinical pharmacology data repository increases the capacity for in vivo PK characterization and is a key component in the IT architecture of modern drug development organizations.
Daniel Weiner, PhD, is chief technology officer, Pharsight Corp.
________________________________
This story first appeared in eCliniqua,one of Bio-IT World’s free e-newsletters. Subscribe here.