Why it is time to make the shift to Data Driven Intelligence to ensure continuous product quality in Life Science

Florian Hildebrand

Florian Hildebrand, Managing Partner, ChemSquare

Growing numbers of batch recalls in Pharma and Life Science, have focused increased attention on supplier qualification and quality control. As globalization and technological innovation advances, challenges of overseeing and regulating suppliers and manufacturers becomes all the more complex.

Health Authorities, like the FDA or EMA suggest periodic audits with a time span of at least every three years for APIs, as part of the supplier qualification process. In addition, pharma producers are every so often required to audit suppliers for excipients, as well as contract manufacturers, laboratories and distributers.

While large pharma companies employ global auditing teams to cope with these ever-increasing audit requirements, SMEs partially have to rely on external partners for regulatory assistance. Thus, outsourcing audits to third party companies has become common practice for SMEs.

As audits are standardized by regulatory guidelines such as EU GMPs (Part I and Part II), they inhibit a “shared-value”. Third party audit companies, therefore, have been taking advantage of these overlaps by offering Shared Audits -meaning a single audit can be shared with several pharma producers- to facilitate the audit process and bring cost efficiency to pharma producers.

However, the full potential of audit-sharing is still being lost by sticking to outdated audit processes. It is time to shift from static-periodic supplier auditing processes to predictive methods with focus on continuity. How can we leverage new technologies such as artificial intelligence (AI) to drive improved product quality and help with early detection of potential risks?

The ChemSquare platform was designed to introduce new ways of interaction between pharma producers, suppliers and auditors. The benefits of digital ecosystems are enormous. Leveraging a digital ecosystem approach and combining it with the advantages of AI and machine learning (ML) could enable Life Science companies to harness data and translate it into valuable insights.

How AI could turn audit reports into concrete value:

The Center for Strategic Supply Research highlights the benefits of supplier ecosystems and further endorses the integration of intelligent algorithms in order to evolve beyond cost reduction towards long-term value creation by increasing regulatory compliance. The use of existing audit reports, from various sources could provide the foundation for initial tests.  But how could that look like?

Deriving data from Audit reports, is at best difficult. But in many cases nearly impossible, as they usually are stored as PDFs or print outs. The problem is that they contain textual data that is unstructured and disorganized, thus cannot be analyzed with traditional IT methods. To be able to analyze these data sets, we require new intelligent ways. Through the application of dedicated AI algorithms, it would be possible to decipher and analyze data from various audits, provided by different sources to identify trends and potential issues with the performance of the entire supply chain. Artificial Intelligence, therefore, could allow for the aggregation of information from these reports, which in turn could provide greater insights into process patterns. The value that lies here is tremendous: It would empower pharma producers in monitoring and tracking supplier input and enable early action before problems even arise.

Where could we start?

Digital ecosystems like the ChemSquare platform can help collecting this information to start an initial analysis. For a first trial it would even be enough to collect less confidential data to test and develop dedicated algorithms.

The next step would require more complex methods to create a standardized audit evaluation process, that AI will be able to reveal relevant information from.

Digital platforms, like ChemSquare, provide the perfect nurturing ground to implement this novel approach and transform the audit process from a disconnected and static state into a continuous-dynamic and connected state. This ultimately could have major business impact by supporting pharma companies in decision-making, planning and responding to issues much quicker. The result: Increased compliance, mitigation of risks and supply-chain optimization.

Pharma producers need to double down on quality

Compliance and Quality are not the same. The question that arises with the batch recall increase is: “Can you distribute poor quality products while complying with the regulatory framework?” The answer is “Yes”. Let’s not delude ourselves and face the facts: Sticking to the minimum of conducting audits every three years as suggested by Health Authorities does not cut it here. Audits are usually conducted to assess the supplier’s conformity to GMP guidelines. But conducted on a regular basis, audits can be used as an effective improvement tool. Batch recalls usually occur because pharma companies are at loss when it comes to auditing their suppliers on a frequent basis. Now is the time to move from outdated quality system models to more dynamically innovative approaches. Once we shift our focus on more predictive methods that are AI driven, we will not only enable pharma producers to be more cost-efficient but more importantly deliver a flexible framework for assessing supplier competencies.

With the help of digital technologies, automation and AI we are able to support quality throughout the whole lifecycle.

What we need now is executive commitment as prerequisite for implementing this novel approach in support of product quality. In Life Science, companies have a laser-sharp focus on Good Manufacturing Practices – which is known for its strenuous process. But with the right tools, supported by technology, pharma companies will be able to leverage relevant manufacturing data to support improved decision making, and thus yielding effective, safe and high-quality products.