Open Source and the FDA

Bringing new therapies - drugs, biologics, and devices - to market in the USA requires FDA approvals.

FDA submissions (documents, code, data), for years, were based on proprietary SAS - but only by convention. In 2015, the FDA made clear that open-source *code* was allowed "provided the sponsor can validate their data management workflows and statistical analyses". Data, based on the international CDISC standards, must still be transferred in a SAS file format but that is opening up to allow XML and JSON data formats.

This presentation is an overview of all the efforts, over the last decade, towards moving away from proprietary SAS to open-source programming languages - primarily R - as well as open data formats - so that open source is the main way that new cures are developed and approved.

We will do a deep dive on the Open Source in Pharma group, the R/Pharma community, the R Consortium, PHUSE, and the pharmaverse work in particular.

We will also look at what efforts have been started regarding Python for Pharma.

In short, this talk will introduce you to how one works with the FDA and to convince you that the future of therapeutic development is open source.