FDA's Evolving Stance on Real-Time Release Testing (RTRT)

For decades, pharmaceutical manufacturers have relied on end-product testing — sampling finished batches and testing them against specifications in the QC laboratory — as the primary assurance that a product meets quality standards before release. This model is well understood, well-established, and expensive to maintain. It is also, as FDA and the broader industry have increasingly recognised, not the most informative approach available when the process itself already contains the quality signal.

Real-Time Release Testing (RTRT) offers a different model: using in-process measurements, supported by process knowledge and validated statistical models, to assure the quality of finished drug product — and to reduce or eliminate specific traditional end-product release tests. FDA's renewed activity in this space signals institutional commitment to RTRT as a pillar of pharmaceutical quality modernisation.

Defining RTRT

ICH Q8(R2) defines Real-Time Release Testing as "the ability to evaluate and ensure the quality of in-process and/or final product based on process data which typically include a valid combination of measured material attributes and process controls."

In practice, RTRT involves substituting a traditional compendial test — dissolution, content uniformity, residual moisture — with a combination of:

• Measured material attributes determined in-process or at-line (e.g., blend uniformity via NIR, API solid-state form via Raman spectroscopy, particle size distribution by laser diffraction)
• Verified process controls (e.g., validated press force and speed settings demonstrated to correlate with in vitro dissolution)
• A proven, statistically validated relationship between the in-process observables and the quality attribute the traditional test was designed to assure

The critical phrase is "valid combination." RTRT is not simply trusting a PAT reading — it is a demonstrated, model-supported relationship between in-process data and the final attribute, with defined boundaries of validity.

The PAT Connection

Process Analytical Technology is the measurement infrastructure from which most RTRT implementations draw. Near-infrared spectroscopy, Raman spectroscopy, acoustic emission sensors, and inline imaging systems can provide real-time, non-destructive measurement of critical quality attributes during manufacturing. FDA's foundational 2004 PAT Guidance opened the regulatory pathway. RTRT builds directly on PAT: where PAT provides the measurement capability, RTRT provides the framework for using that capability in lieu of traditional end-product testing.

This distinction matters operationally. Many manufacturers already operate PAT instruments for process monitoring and in-process control. The step to RTRT involves formalising the model, validating the surrogate relationship, and submitting it to regulators as a release mechanism — a meaningful additional commitment, but one that builds directly on existing investment.

What FDA Now Expects

FDA's current expectations for RTRT applications, as reflected in Quality Modernization communications and ICH Q8-aligned guidance, centre on several core requirements:

Model development and validation: Any surrogate measurement underpinning RTRT must be supported by a chemometric or statistical model developed using a representative sample set, validated against independent samples, and demonstrated to be robust to normal process variation. The model is not a static equation — it requires formal monitoring, defined requalification triggers, and periodic re-validation as raw material sources, equipment, or process conditions evolve.

Design space integration: RTRT functions most reliably within a defined design space (ICH Q8) where process variables are controlled within validated ranges. If the process operates outside that space — due to raw material variability, equipment deviation, or process drift — the validity of the RTRT model may no longer hold. Clear criteria for when traditional testing must be invoked as a fallback are expected.

Regulatory submission pathway: RTRT must be comprehensively described in the regulatory filing. For NDA and ANDA applications, this means a CMC section describing the RTRT strategy, measurement methodology, model development and validation data, ongoing monitoring programme, and explicit identification of which traditional tests are replaced, reduced in frequency, or retained.

Risk-based retention of traditional testing: FDA does not expect complete elimination of traditional testing in most RTRT strategies. Rather, RTRT replaces specific tests while others remain, or reduces the frequency of testing for stability of the approach. The allocation of testing responsibility — what RTRT covers, what traditional testing still covers — must be explicitly risk-justified in the submission.

Business Case and Operational Advantages

The business case for RTRT extends well beyond laboratory headcount reduction. Organisations that have implemented RTRT report several structural benefits:

• Reduced batch release cycle time: Eliminating a 45-day dissolution test as a release criterion, for example, directly compresses time to market. For high-demand products or supply-constrained situations, this can have material commercial impact.
• Real-time process intervention capability: PAT data underpinning RTRT provides continuous process insight, enabling manufacturing to identify and address variability before it produces out-of-specification results — shifting from reactive investigation to proactive control.
• Continuous manufacturing enablement: RTRT is functionally prerequisite for continuous manufacturing platforms, where traditional batch-based sampling does not map cleanly onto the continuous process unit operations.
• Deepened product knowledge: The extensive characterisation required to validate an RTRT model — establishing the surrogate relationship, defining model boundaries, stress-testing robustness — generates product and process knowledge that pays dividends in regulatory interactions, deviation investigations, and lifecycle management for the full commercial lifetime of the product.

Honest Limitations

RTRT is not the right approach for every product or every release test. Several practical challenges warrant candour before embarking on a programme:

High development cost: The investment in PAT hardware, chemometric model development, validation studies, and regulatory preparation is significant. For smaller organisations, low-volume products, or tests where the surrogate relationship is poorly understood, the return may not justify the approach.

Model maintenance burden: Raw material variability, equipment changes, and process drift can all invalidate an RTRT model that performed well at validation. A robust model monitoring programme — with defined acceptance criteria for ongoing predictions, periodic re-calibration, and clear requalification triggers — is mandatory and is consistently underestimated in initial project scoping.

Multi-market regulatory alignment: While FDA and EMA have both signalled support for RTRT, specific expectations around model complexity, validation batch numbers, in-market monitoring, and acceptable model update procedures vary by market. Global organisations must plan RTRT strategies with a multi-agency perspective from the outset.

A Practical Path Forward

For organisations considering RTRT, the most productive starting point is identifying a single high-impact test where the surrogate relationship is well understood, PAT capability already exists, and the regulatory risk of a novel approach is manageable within the organisation's current regulatory relationship with the relevant agency.

Water content assayed by Karl Fischer titration is a natural first RTRT candidate for solid dosage forms where NIR moisture measurement is already in use: the in-process measurement is well characterised, the surrogate model is relatively straightforward to validate, and the regulatory precedent is growing. From that first application, the organisation builds measurement capability, regulatory confidence, and internal competency in chemometric model management — creating the foundation to expand RTRT to more complex critical quality attributes and additional products over time.

FDA's Quality Modernization initiative reflects a long-term strategic direction, not a short-term programme. For pharmaceutical manufacturers committed to building quality systems that are genuinely predictive rather than retrospective — systems that detect problems before they occur rather than confirming compliance after the fact — RTRT is not a distant aspiration. It is a near-term strategic priority worth investing in now.