Process understanding
The success of any particular technology transfer depends upon process understanding, or the ability to predict accurately
the future performance of a process. It requires complete, methodical, and systematic capture and codification of all process
data and information leading to a high level of assurance regarding site-to-site process equivalency. In the case of transfer
from development to commercial manufacturing, the goal of process equivalency is complicated by the issues of scale-up—what
works in the laboratory may not work in the plant.
Figure 2: Flattening the learning curve—demonstrating that time to market and improved knowledge capture is faster with an
integrated evelopment/commercialization team approach.
|
To forestall such problems, development and commercial teams should interact early in the process of development. This involvement
of manufacturing personnel during development enables development staff to understand scale-up limitations and design appropriate
experiments, thus saving time. Conversely, the involvement of development personnel on scale-up teams leads to better management
of knowledge transfer. Working together, development and manufacturing personnel can continue to build a broader understanding
of what is and is not practical for commercial manufacturing. The result is more intelligent and efficient development activity
and a flattening of the learning curve as products move from development to pilot to commercial operations (see Figure 2).
Unfortunately, in many companies, such interaction does not begin until after successful Phase III trials, in the belief that
it is too costly before that stage of development because so many drugs fail. But these same companies then find themselves
with little time for achieving sufficient process understanding for successful scale-up. As a result, a process that is less
than robust is transferred to manufacturing and encounters time-consuming problems that delay the product's launch. Thus the
penny-wise and dollar-foolish refusal to engage manufacturing with development sooner often results in missed revenue, regulatory
issues, and nagging production problems that continue to crop up and incur otherwise avoidable costs. In fact, a number of
companies that have learned this lesson the hard way are now applying the tools of process understanding much earlier in development
to ensure efficient manufacturability later.
The following three elements help achieve this all-important process understanding:
- Using quality-by-design (QbD) techniques and identifying critical process parameters (CPPs), which are those process parameters
that must be maintained within critical limits to achieve the desired quality outcome.
- Identifying drivers of variation and of interactions in critical processes
- Understanding the right process-control strategies.
QbD establishes a product's design space, which comprises the multivariate relationships that encompass combinations of product
design, manufacturing-process design, manufacturing-process parameters, and raw-material quality that ensure suitable quality
and performance. By establishing design space in the initial development of a product, the organization gets a head start
toward successful transfer to manufacturing. CPPs will have been identified, as well as the drivers of variation and interactions
in critical processes. The organization will know how the process will behave under various multivariate conditions, such
as are likely to occur at commercial scale.
The goal for products that weren't developed through QbD and are being transferred from one site to another is, in effect,
to achieve a similar level of process understanding retroactively. For example, a product may have been put into production
at the initial site without sufficient process characterization before validation, and some of the CPPs might not have been
identified. To fill in those blanks, it is advisable to begin by developing qualitative process understanding through focus-group
interviews, process maps, and document review. It is also necessary to conduct basic quantitative characterization of the
process, using statistical tools such as control charts and CpKs (a statistical measurement of process capability) to provide
a quantitative assessment of whether a process parameter is capable of meeting its specification limits. Advanced statistical
tools, such as regression analysis and design of experiments (DoE) can help establish the causal relationships and understand
interactions between a finished product's release parameters and other in-process parameters and raw-material characteristics.
The level of process understanding necessary for successful transfer has been achieved when the impact of changes in critical
process inputs can be predicted. Process-control problems that might occur in scale-up or site-to-site transfer can thereby
be anticipated and preempted. Furthermore, process understanding means that all critical sources of potential variability
(and interactions) have been identified and explained, and product specifications are based on understanding of sources of
variation and process capability. With this understanding, it is possible to make sure that the transferred process will operate
at the center of the multivariate relationships—the operating sweet spot where process predictability and robustness are found.
Process robustness is a critical success factor for right-first-time transfers that minimize impact on the supply chain. Benefits
of these on-time, on-budget transfers include the following items:
- Reliable supply to market
- Higher product or process reliability
- High-quality products with few to no defects, loss, or rework
- Reduced operational and compliance risk
- Lower product and operational costs
- Higher product development-to-launch success rates
- Faster overall time to market.
Together, a carefully constructed technology-transfer framework and greater process understanding will not only help an organization
achieve these benefits, but also make sure that it can be done time and time again.
William Schmidt* is a principal and pharmaceutical sector head, and Ian Uydess is a managing consultant and project manager, both at Tunnell Consulting, 900 East Eighth Avenue, Suite 106, King of Prussia,
PA 19406, tel. 610.337.0820, schmidt@tunnellconsulting.com
.
*To whom all correspondence should be addressed.
|
