Common Considerations for Submitting Pharmacokinetic (PK) Values in CDISC Format
02 Feb 2018
Any human clinical trial generates pharmacokinetic (PK) data, and those are the basic raw data for statistical analysis where BA/BE parameters are the primary endpoints. PK data are also used to determine dosing regimen, dose tolerability, and effective dosing interval.
Here, we discuss some of the challenges associated with representation of the collected concentration data and the calculation of PK parameters followed by conversion in the Clinical Data Interchange Standards Consortium (CDISC) compliant format.
CDISC Background – CDISC is a global organization that develops standards to support the acquisition, tabulation, and submission of clinical data. The mission of CDISC is ‘to develop independent data standards that enable informality for medical research’. More information is available at www.cdisc.org
SDTM Background – The Study Data Tabulation Model (STDM) is a broadly used CDISC standard to submit clinical data in a well organized format. SDTM Implementation Guide (SDTMIG) describes mainly two SDTM domains for PK representing pharmacokinetic concentrations (PC) and pharmacokinetic parameters (PP) datasets; additionally Related Records (RELREC) domain is used for the presentation of the relationship between PC and PP domains. It is also important to incorporate CDISC controlled terminology for PK domains.
PC domain is used for the tabulation of all concentration results for all possible samples (plasma, urine) and has a structure like one record per subject per dosing per drug per time (date and time); therefore in PC domain, subject profiles are represented across multiple records. The PC domain is the base for pharmacokinetic analysis, and can be used for the calculation of parameters like Cmax, Tmax, AUC, etc. which can be further submitted in the PP domain.
Relating PC and PP Records – Once the SDTM datasets are ready, there is one more important thing to do, while submitting the PK data in CDISC format, and that is linking all derived PP records back to the original PC observations. Creation of RELREC is the primary practice to establish relation between PC and PP domains. Basic concept of RELREC is that the related records from each domain (PC & PP) are presented in nearby rows using SDTM standard identifier variables (IDVAR, IDVARVAL, and RELID). RELID is used to uniquely identify records within the dataset.
Lambda Therapeutic Research has rich experience in PK analysis and creation of CDISC compliant datasets for regulatory submissions. Recently, we also helped our sponsor (A pharmaceutical client for NDA submission) to re-submit their legacy data in CDISC format to FDA in response to a regulatory query.