The clinical trials industry might get an XPT alternative soon in dataset-JSON. The new dataset-JSON format is being seriously considered to be the default file format for regulatory submissions.
While most of the aspects of the regulatory submission process in clinical trials have evolved over the years, the SAS V5 XPORT (XPT) requirement remained the same. Now submitting study data sets for regulatory submissions is going to take a new form.
The Clinical Data Interchange Standards Consortium (CDISC) and PHUSE, the global healthcare data science community, are joining hands with FDA and leading the way for a pilot project that aims to bring about a change in the existing dataset exchange format in clinical trials. The new Dataset-JSON is developed by CDISC and aims to replace XPT as the default file format in the coming future.
Being conducted in two phases (Short-term and Development of Future Strategy), the objective of the FDA pilot project is to support the adoption of Dataset-JSON as a substitute for transfer format in clinical and device data submissions for regulatory authorities and other dataset exchange instances. While the goal of the pilot’s first phase is to ascertain Dataset-JSON as a drop-in replacement for XPT, the second phase will focus on the long-term purpose of further fortifying the new versions of the CDISC Foundational Standards.
The pilot program will also put an emphasis on the usage of Dataset-JSON with the existing processes to add it to the Data Standards Catalog and make it an accepted standard for regulatory submissions.
Well, why did the need arise now for the globally-accepted XPT format to be replaced?
Dating back as far as 1989, XPT has become a standard format for representing datasets in the CDISC Foundational Standards. But it is slowly becoming outdated and troublesome with the changes in the clinical technology landscape. Being a binary format, its applications for data exchange in this digital age are very finite.
Moreover, XPT is proving to be quite difficult with a lot of limitations when it comes to data submission and is riddled with hassles such as inefficient storage utilization, limited data types, insufficient metadata, field name and variable size challenges, no Unicode support, and no extensibility.
In order to overcome all these hurdles, CDISC developed the Dataset-JSON, which will be a part of its new ODM V2.0 standard, based on the CDISC Dataset-XML V1.0. It supports a variety of programming languages and technology stacks. What’s more interesting is that a 2022 Dataset-JSON hackathon has resulted in the development of a multitude of open-source software tools to easily convert Dataset-JSON to Python data frames, SAS datasets, and R data frames.
So how will Dataset-JSON become mainstream in clinical trials?
It all depends on the success of the pilot project to deduce the acceptability of Dataset-JSON by the FDA and the industry heavyweights. If the industry is to veer more towards the use of Dataset-JSON, there are two certain prerequisites:
a). The study dataset submission process should not be affected or disturbed at any cost as the industry moves to utilizing Dataset-JSON and
b). The regulatory authorities completely trust Dataset-JSON over the XPT.
The hackathons, the workshops, and the FDA pilot are aimed at ensuring that the prerequisites are met for Dataset-JSON to become the standard format for data submission in clinical trials.
Stay tuned to our social media and upcoming blogs for more such interesting and timely information around life sciences and clinical trials.