Submitting Real World Data is a project within the PHUSE Real World Evidence Working Group. Since the inception of this project, we’ve had several meetings to discuss the challenges faced by industry while dealing with submission of real-world data (RWD) to regulatory bodies. There are several guidance documents released by regulatory bodies (especially the US FDA and the EMA) with regards to handling and management of real-world data. (1,2) Some of these guidance documents are still in draft stage and some are considered final.
From the feedback received from this project’s members, there is ambiguity and uncertainty about how to deal with real-world data, especially when it comes from different data sources in general clinical trial settings. Unlike the data collected through randomised controlled clinical trials, real-world data poses challenges to source data verification, which in turn results in data traceability, bias, validity and provenance-related issues. As a general theme, these issues impact on how the data can be submitted to regulatory bodies, especially when the general position of regulatory bodies like the US FDA is about submission of such data in CDISC standards, (4) unless otherwise discussed and agreed between the FDA and the sponsor.
Looking further into details, the challenges faced by this project vary depending on data source and location of collected data. The general theme of the challenge is what and how to submit real-world data.
What to Submit – Compliance Issues
In real-world studies, the source data is processed, curated and transformed before it is used in clinical trials. So, the biggest challenge for submission of such data is about what stage of data can be considered true source data. Moreover, since the source data is owned and managed by a data vendor, unless clearly specified in the contract, the vendor may not provide source data given the patient data privacy issues. There is a similar challenge in getting such data from a country where the source data privacy-specific regulations are strict. In such circumstances, it is very challenging to explain and submit the source data to regulatory bodies.
In addition, there is a varied degree of data anonymisation used, as per country regulations. Getting one agreeable degree of data anonymisation can be challenging. Some regulatory bodies also deal with other healthcare data standards. Since the majority of RWD comes from healthcare data sources, some sponsor members have experienced regulatory bodies accepting certain RWD not in the CDISC standards. Such submission of data was accepted only with prior communication about usage of the data with the concerned regulatory bodies. So, based on interpretation of the regulatory guidance available, and on the experiences of member representative companies, CDISC is still a recommended standard for submission of RWD, though other formats may be accepted if discussed with the agency in advance of submission. Sponsors need to communicate with the relevant regulatory body in the early stages of clinical trial planning on any exclusions for a CDISC-based data submission.
How to Submit – Compliance Issues
Given that the RWD is collected in general medical settings, such data does not necessarily have information correlating to all required and expected SDTM variables. As a result, adherence to CDISC SDTM, and compliance of SDTM, becomes very challenging. Another important aspect discussed in our meetings includes getting baseline information from RWD when it is scattered among data elements. Given the various data sources of RWD, there is a lot of additional data which does not fit well into CDISC SDTM. It is unclear to many of our group members how such data can be submitted to regulatory bodies. One FDA guidance mentions that sponsors may include an appendix of RWD-specific elements/variables in data reviewer’s guides. However, expectations about what needs to be included in an appendix or addendum is left up to the sponsor to decide. Some RWD submissions done by sponsor members involve planning and prior discussion with regulatory bodies. Depending on the negotiations between the regulatory body and the sponsor, how and what data to submit is decided. Consequently, as of now, there does not appear to be one consistent approach with regard to how to submit RWD to regulatory bodies.
PHUSE Project Efforts
There is certainly an opportunity to streamline aspects of submission of real-world data to regulatory bodies. The Submitting Real World Data Working Group project is focused on interpretation of regulatory guidance and development of industry-wide common processes which can help industry and regulatory bodies in submission and communication of real-world data. Our core focus is to make sure there are streamlined processes and documentation collaterals which meet necessary as well as sufficient data submission requirements of regulatory bodies and facilitate the review of real-world evidence submissions. Our project’s current efforts are focused on finalising a checklist for RWE trials. Further efforts are planned for development of a template for a specific appendix in the data reviewer’s guide for RWE studies. A by-product of these efforts could result in further proposals to other industry bodies for collaborative initiatives.
References
- Data Standards for Drug and Biological Product Submissions Containing Real-World Data | FDA
- Considerations for the Use of Real-World Data and Real-World Evidence To Support Regulatory Decision-Making for Drug and Biological Products | FDA
- real-world-evidence-framework-support-eu-regulatory-decision-making-report-experience-gained_en.pdf (europa.eu)
- All data standards development efforts from the Clinical Data Interchange Standards Consortium can be found at CDISC | Clear Data. Clear Impact.
Authors: Parag Shiralkar, Matt Baldwin, Lorraine Fang, Joe Xie, Mary Anne Rutkowski, Sandy VanPelt Ngyuyen, and members of the Submitting Real World Data Working Group project.