Process validation across the lifecycle: the three stages

For a long time, process validation was considered done as soon as three consecutive batches met the specification. This mindset persists stubbornly, even though the regulatory framework moved beyond it long ago. Today, process validation is a lifecycle that begins with development and continues throughout commercial manufacturing. Treating it as a formality before market launch squanders its real value: a process that stays predictable.

From batch thinking to lifecycle

The FDA guidance Process Validation: General Principles and Practices from 2011 set the framework that the EMA carries in parallel in Annex 15 and its process validation guideline. Instead of a one-time hurdle, it describes three interlocking stages. This is underpinned by the ICH guidelines Q8 to Q12, which bring together pharmaceutical development, risk management, quality system and lifecycle management.

Stage 1: Process Design

In the first stage, the process is understood, not merely executed. Based on development and risk assessment, the product's critical quality attributes (CQA) and the critical process parameters (CPP) that influence those attributes are determined. ICH Q8 refers to the Design Space, the range within which the process is demonstrably permitted to operate. What is cleanly characterised here does not have to be laboriously tested out later, and what is missing here resurfaces as a surprise at scale.

Stage 2: Process Qualification

Only in the second stage is the commercial process formally confirmed. Process Performance Qualification (PPQ) demonstrates that the process reproducibly delivers what it should at the target scale and under routine conditions. This requires that equipment, utility systems and analytical methods are already qualified. The widespread figure of three batches is not a law here: Annex 15 and the FDA require a risk-based, justified number derived from process complexity, experience and variability.

• Foundation: a predefined validation plan with clear acceptance criteria.
• Evidence: reproducibility across the planned batches, not just conformity to specification.
• Result: a process that is demonstrably under control, not merely successful once.

Stage 3: Continued Process Verification

The third stage is the one most often underestimated. Continued Process Verification (CPV) keeps the process under observation throughout its entire lifecycle: routine data are continuously trended statistically in order to detect drift, special causes and gradual deterioration before they turn into a deviation. Here, process analytical technology (PAT) and statistical process control provide the tools to turn raw data into genuine early warning.

A validated process is not a state but a proof that is continually renewed.

Traditional, continuous or hybrid

For the validation itself, the EMA permits three routes: the traditional approach with a defined number of batches, Continuous Process Verification, which derives the evidence directly from comprehensively captured process data, and a hybrid approach combining both. Which route fits depends on process understanding and the available data basis, not on convenience.

What counts in the end

The typical weak point lies at the transitions: if the critical parameters from Stage 1 are not cleanly translated into the PPQ criteria of Stage 2, the testing misses the actual risk. And if Stage 3 sits on data that are collected but never evaluated, a deceptive sense of security arises. Process validation thrives on the chain, not on individual links. Vispact supports such validation projects along the entire lifecycle, from parameter definition through to statistical trending in ongoing operation.