According to PDS, SEND Express is a “turnkey solution for the generation of Standard for Exchange of Nonclinical Data (SEND) datasets.”
"With SEND Express, PDS’ SEND experts work as a virtual extension of our clients’ team," Sayed Badrawi, CEO, told Outsourcing-Pharma.com. "We aggregate and harmonize data from multiple organizations, information systems and file formats ... to produce one set of harmonized SEND files, including define.xml and define.pdf, as required by the FDA."
SEND is an implementation of the CDISC Standard Data Tabulation Model (SDTM), which specifies a way to present nonclinical data in a consistent file format, as well as variables and standardized scientific terminology.
To convert a client’s data into SEND-compliant datasets, PDS’ preclinical scientists and data analysts use TranSEND, the company’s proprietary software data translation engine.
"Throughout the dataset generation process, multiple quality control steps are completed to ensure the integrity of the datasets," added Badrawi.
"Furthermore, this work is supervised by our veteran toxicologists and compliance experts who are active within the PhUSE and CDISC communities and current with preferred practices, recommendations and requirements."
Setting the new standard
The platform’s launch parallels the FDA’s deadline for standardized electronic data submissions, which will require all datasets from studies supporting new drug applications, biologics license applications, and abbreviated new drug applications to be submitted in SEND format by Dec. 17, 2016.
The SEND package includes several components, such as endpoint and study design data in the form of domain datasets and a description of the dataset provided in define files.
As Badrawi explained, "Within the standard, the variables for each domain dataset are specified to ensure consistency." Additionally, the FDA recommends including a Study Data Reviewer’s Guide with each SEND package, yet this is not required by the standard.
According to Badrawi, the implementation of SEND standard is "exceptionally challenging." Specifically, "because in preclinical research, much of the vital data is based on anatomical organ pathology findings and those findings are recorded based on glossary terms that vary by pathologist," he explained. Because of this, study data must always be “mapped” to SEND-specified controlled terminology before the SEND transformation is initiated.
"This is a complex process and requires the input of preclinical research scientists," added Badrawi.
However, "Having an industrywide, harmonized model allows for efficient analysis by regulatory authorities," he added. "The common formats assist in enabling the exchange of nonclinical data within companies and consortiums. Additionally, this consistency of data makes it easier for vendors to develop tools that can use this data."