Walk into our analytical development and stability laboratory in Verona on any morning, and the first thing you notice is what isn’t there. There is no stack of paper batch records on the bench, no clipboard hanging from a stability chamber, no analyst hunched over a notebook copying numbers from an instrument screen. The high-performance liquid chromatography (HPLC) system is already logged in. The scientist at the bench scans a sample identifier into a tablet and within seconds the test plan is on screen. The standards and samples are queued and sent to ‘Empower’, and the data, when it arrives, will land directly into our laboratory information management system (LIMS) without a single line being typed by hand.
This is not how most analytical laboratories work today. At Evotec, we have worked in a digital environment for years. But now we have taken our level of digitalisation to a further level. One single system covers all processes, from:
• sample logging to reports generation
• Instruments
• Calibrations
• Stability chambers
• Pull samples
• Investigations
• Log retest samples – all in one.
The way regulators are asking the industry to submit pharmaceutical quality data is rapidly changing, and that change is forcing every CDMO and big pharma to fundamentally redesign their data infrastructure from the bench upward.
Financial industry is a benchmark
The financial industry provides a useful benchmark. Equally regulated, it runs on a fully digital, interoperable ecosystem where transactions clear in real time, audit trails are inherent and data integrity is structural. Pharma, similarly regulated and data-driven, has been slower to get there. That gap is no longer sustainable.
Good news is that the US Food and Drug Administration has been moving steadily toward structured, machine-readable submissions. Two initiatives are the most visible:
- PQ/CMC (Pharmaceutical Quality / Chemistry, Manufacturing and Controls), which standardises electronic CMC data using HL7 FHIR; and
- KASA (Knowledge-Aided Assessment and Structured Application), which layers AI and structured assessment on top.
The implication for any laboratory generating regulatory data is concrete. The narrative-and-PDF model of dossier preparation is on the way out. Replacing it, is data that flows from the instrument to the agency, traceably, in a structured format.
We saw this coming early, in part because one of our big pharma clients put it on the table directly. They were preparing for the new submission standards and asked us a clear question: ‘when our samples come out of your stability programme, can the data flow into our repository in the structured format the regulator will expect, without anyone retyping it?’ We agreed it needed to be done, and we agreed it had to be a feature of how we operate, not a one-off integration for one project.
What we built together, over many months, is now in routine use across our analytical development and stability laboratories. We call it the Next Generation Analytical Data Flow Interface, or NxADFI for short. It sits between our LabWare-based PharmaLIMS environment and the client’s data systems. Instruments at the bench, including balances, pH meters, HPLC and UPLC systems running Empower, feed PharmaLIMS automatically. NxADFI, in turn, extracts these data on demand and transforms them into machine-readable format such as JavaScript Object Notation (JSON) and potentially others (eXtensible Markup Language (XML) or via Fast Healthcare Interoperability Resources (FHIR)-compliant APIs). The output data are structured exactly the way each client needs to absorb it. The transcription steps on the client side, which once demanded hours of effort and posed the greatest risk to data integrity, have been eliminated.
A typical busy day
What does that look like in a typical day? On a busy stability time-point, a single analyst can release results in a fraction of the time it used to take to manage the same volume on paper. Certificates of Analysis and stability reports are generated automatically from validated templates. If a result trends out of expectation, the out of specification (OOS) or out of trend (OOT) investigation workflow is already in the system: no separate paperwork, no scanned signatures and no version control headaches. Approvals happen electronically with full audit trails. If a colleague in another country needs to review a result on a tablet, they can. Every action is captured.
The shift is not only technical. It changes what an analyst’s day looks like. Less time is spent transcribing, reconciling and chasing missing initials; more time is spent doing the science: investigating an interesting peak, optimising a method, mentoring a junior colleague. Quality assurance reviews focus on the integrity of the science rather than the integrity of the paperwork.
For a CDMO running dozens of programmes in parallel, that compression of cycle time has real consequences for our partners’ timelines.
IT systems to work alongside the client
Behind all of this sits something less visible but equally important: a validated computer system, built and maintained to GMP. Our IT, computer system validation and analytical teams work alongside the client’s scientific and IT teams from the start of any programme. The interface itself is encrypted and uniquely tagged per client and per study, so confidentiality is structural, not procedural. Every step generates logs that are placed in an immutable vault.
Conclusion
When PQ/CMC becomes the standard rather than the direction of travel, and it will, the laboratories that have already restructured around it will be the ones ready to support their partners through it. That work has been done here in partnership with the clients who pushed us to do it. More than any individual technology, it is this fully connected, end-to-end digital way of working that has fundamentally changed how we see and run our laboratories.
