A maze

Pitfalls and promises of open-source software in pharma data analytics

In the fast-evolving landscape of pharmaceutical data analytics, the allure of open-source software is undeniable.

The promise of cost-effectiveness, innovation and collaborative development has led many in the industry to explore the integration of open-source tools into their data analysis workflows.

Regulatory pitfalls, however, can arise when navigating the intersection of open-source software and compliance with regulatory bodies such as the European Medicines Agency (EMA) and the US Food and Drug Administration (FDA).

When conducting clinical data analysis using an open-source package, care needs to be taken to ensure the process meets stringent regulatory standards set by the authorities.

The EMA and the FDA, two major regulatory bodies, play pivotal roles in ensuring the safety and efficacy of pharmaceutical products. From a regulatory perspective, the use of open-source software introduces a myriad of challenges that must be carefully considered.

The FDA’s stance on statistical software is clear – while there is no explicit requirement to use specific software, the choice should be fully documented in submissions, including version and build identification.

The FDA emphasises the importance of reliable software for data management and statistical analysis, urging sponsors to consult with FDA statisticians early in the product development process.

Similarly, the EMA requires adherence to standards such as the Clinical Data Interchange Standards Consortium (CDISC), specifying the format and organisation of study data. Any deviation from these standards may pose a significant hurdle in the regulatory approval process.

Recently, there have been documented incidents with pharmaceutical companies that underscore the real and perceived challenges associated with open-source software.

Independent vendor software is often viewed as more stable, with legacy code and dedicated technical support providing a sense of security. Open-source solutions, on the other hand, may face scepticism regarding validation, support and potential intellectual property issues.

To address these challenges, a hybrid approach has been proposed. By integrating open-source tools into a secure environment and leveraging proprietary software for specific purposes, companies can harness the benefits of both worlds.

Stability concerns can be mitigated through well-managed package systems, and legacy code can be accommodated through hybrid workflows.

Solutions to challenges include adapting IT departments to support open-source software, using package archives for distribution and addressing intellectual property concerns through legal assessments.

The industry should consider these solutions to pave the way for a smoother integration of open-source software in the regulatory landscape.

While challenges exist, open-source software also presents opportunities for cost savings, innovation and enhanced data sharing.

Recent graduates are often more familiar with open-source tools, and the wealth of available packages can drive interactive data visualisations and dashboards. It’s important that the industry is able to explore these advantages while remaining vigilant about regulatory compliance.

As we navigate the dynamic intersection of open-source software and pharmaceutical data analytics, a cautious and informed approach is key.

By understanding the regulatory landscape and adopting a hybrid strategy, companies can harness the power of open source in a secure and compliant manner, ensuring the integrity of clinical data analysis and regulatory submissions.

Greg Wujek is Manager Life Sciences Industry Consulting at SAS.
Go to sas.com