Anne Kümmel (1), Henning Schmidt (1)
1: IntiQuan GmbH, Switzerland
Objectives: Non-compartmental analysis (NCA) is a well-established approach to quickly gain valuable insights into a drug’s pharmacokinetics as well as to assess bioavailability, bioequivalence and dose-proportionality given the appropriate data. NCA is often performed by dedicated scientists using proprietary software such as Phoenix WinNonlin [1]. Several open-source alternatives are available (e.g., [2,3,4,5]), these however often lack inbuilt functionality to easily perform NCA with flexible options around the computation of the area under the curve (AUC), or the handling of data below the limit of quantification (BLQ). Finally, data programming, data exploration and reporting of the results are usually not covered and require additional code to be written.
Methods: IQnca was designed to provide functionality supporting the whole workflow for conducting an NCA comprising
- data preparation from CDISC or WinNonlin datasets,
- exploratory data analysis,
- NCA conduction, and
- support of regulatory-compliant PK listings, tables, and figures.
For traceability, an data format was established storing all settings for data handling in tables, summary and for NCA conduction. Functionality was implemented to produce informative graphics supporting decisions on outlier handling and appropriate terminal slope calculation. That is, each individual profile is displayed with key meta data (e.g., subject identifier, dosing, profile type, etc.) and color coded highlighting of data issues (e.g., outliers, poor slope determination). Default NCA settings and reporting standards were set according to recommended methodology [6]. However, these settings, e.g., BLQ handling or AUC calculation methods can be customized by the user. Finally, PK concentration and parameter listings and tables can be generated in a format suitable for regulatory submission. For generation of standardized regulatory-report listings, the package IQReport is utilized that can automatically generate Word files containing the listings, tables, and figures. The IQnca outputs have been validated against proprietary NCA software. The package and its use was documented using Rbookdown.
Results: IQnca was published as an open source R package (https://iqnca.intiquan.com/) providing functionality for a reproducible and regulatory-compliant NCA in a script-based workflow. A workflow consists of the data import from potentially CDISC datasets and data exploration that is designed to support the decision making on appropriate data handling and NCA settings. The settings are applied and stored in the amended NCA dataset. The subsequent calculation of PK parameters and generation of listing, tables and figures are then automated to high degree using a limited set of function calls. In combination with IQReport, regulatory compliant Word documents for listings, tables and figures can be easily produced within R.
Conclusions: The functionality of IQnca covers all activities usually required for an NCA – from data import to reporting. While standard approaches are implemented as defaults it provided flexibility to customize data handling and NCA calculation methods. Thus, IQnca overcomes many of the limitation of the previously available R packages for NCA.
Its free availability, comprehensive documentation, and the wide-spread use of R in the pharmacometric community gives the potential for a broad use and user-friendly conduct of NCA using IQnca. Future extensions will include the statistical analysis of the NCA results, e.g., statistical test for dose-proportionality or bioequivalence.
References:
[1] Phoenix® WinNonlin® (Certara L.P. (Pharsight), St. Louis, MO)
[2] C. Acharya et al., A diagnostic tool for population models using non-compartmental analysis: The ncappc functionality for R. Computer Methods and Programs in Biomedicine. 2016, 127:83-93
[3] W. S. Denney et al., Simple, Automatic Noncompartmental Analysis: The PKNCA R Package. Journal of Pharmacokinetics and Pharmacodynamics. 2015, 42(1):11-107,S65
[4] K.-S. Bae, NonCompart: Noncompartmental Analysis for Pharmacokinetic Data. R package version 0.4.7
[5] T. Jaki, M. J. Wolfsegger, Estimation of pharmacokinetic parameters with the R package PK. Pharmaceutical statistics. 2011, 10(3):284-288
[6] PhUSE CSS Development of Standard Scripts for Analysis and Programming Working Group. Analyses and Displays Associated to Non-Compartmental Pharmacokinetics – With a Focus on Clinical Trials. White paper Version 1.0, 25 March 2014
[7] IntiQuan, IQReport: Supporting Efficient Word Report Generation. 2020. https://iqreport.intiquan.com/
Reference: PAGE 29 (2021) Abstr 9660 [www.page-meeting.org/?abstract=9660]
Poster: Methodology - Other topics