Challenges in Analytical Method Transfer

AAPS, San Antonio (Oct. 31)-Industry and regulatory agency concerns over process variability have prompted both groups to take a closer look at analytical method transfer processes. During the presentation “Challenges in Analytical Method Transfer” at the AAPS Annual Meeting, Horacio Pappa, PhD, scientist and Latin American liaison, Standards Development, US Pharmacopeia (Rockville, MD, www.usp.org) briefly discussed why USP is interested in working with analytical method transfers (AMTs).

Pappa presented the three stages of method life cycle; namely, R&D method development, quality control (QC) methods used in QC labs, and QC USP methods used in generics. He explained that USP is introducing Chapter <1226> “Verification of Compendial Procedures” that would apply between the last two stages for how to verify suitability for “intended use.” Pappa stated that USP has started initial discussions of a proposed chapter applicable between the first two stages (after R&D method development) for the transfer of analytical procedures from lab to lab and encouraged attendees to submit their thoughts on this proposed chapter.

AMTs explained

Saji K. Thomas, associate director of quality operations at Par Pharmaceuticals (Spring Valley, NY) reviewed key aspects of AMTs, emphasizing the importance of timing and the preparedness of both the “sending” and the “receiving” companies. Defining method transfer as “the process that qualifies a lab to use an analytical test,” Thomas outlined the four main reasons why an analytical method may be transferred: It is part of the natural progression of the product’s life cycle; to provide additional capacity; the business has relocated; and as a result of corporate merger and consolidation.

Thomas also reviewed a list of “criteria for success,” including the skills of the individual assigned to the project, defining the roles and responsibilities between the sending and receiving units, and the receiving units’ ability to manage expectations and handle inevitable deviations during implementation. “The most important part of the process is the transfer of knowledge gained during the lifecycle of the method, and the key is that the sending unit and receiving unit agree on the acceptance criteria” said Thomas. Acceptance criteria should be established before the transfer.

One increasing trend is the “covalidation approach,” which involves the receiving lab in the validation process and in identifying the validation parameters that must be challenged. Another approach is for the receiving lab to file a method transfer waiver, which may be applicable when the lab is already familiar with the procedure, when the process is a line extension, when its analytical methods are the same or very similar to those in use, when the person who developed and validated the method moves to the receiving lab, and when the new method does not introduce significant alterations. “Most important, CGMP-released material should not be used for analytical transfer because of the risk of a recall situation. Labs should use experimental or expired lots, preferably with a Certificate of Analysis,” said Thomas.

To facilitate documentation, templates should be used and all chromatograms should be attached. In particular, “Make sure the receiving lab can meet all limit-of-quantitation criteria and can obtain a similar impurity profile. Spiked samples could be used if impurities are not detected above the reporting threshold. This is the only way the performance of the impurity method can be verified,” said Thomas. For dissolution methods, he suggested using Level 2 testing (12 units) instead of the 6 units typically used for method transfer. In addition, bracketing may be appropriate for multiple strengths.

Over the lifecycle of the product, the method may have to be revalidated, especially when the sponsor makes changes to the analytical procedure. “The need to transfer a revalidated method depends on the severity of the change. The receiving unit can perform the method only if the method is robust,” said Thomas.

Transferring particle size and other nonchromatographic methods

Dana C. Vaughn, principal researcher at Procter & Gamble Pharmaceuticals (Norwich, NY) presented common pitfalls pharmaceutical companies encounter when transferring nonchromatographic methods. “Method transfer enables the production of relevant and meaningful data at a satellite facility. The purpose of method transfer is to define quality,” he said. Particle size distribution (PSD) presents unique challenges because there is little guidance on PSD methods. Problems also stem from sampling procedures (because the mean particle size is changed when a sample is pulled) as well as from the instrument response functions that highlight variation in detector geometry and scattered light intensities. PSD methods may also involve proprietary algorithms, and setting particle size specifications is not easy because it may have stability, processing, and content uniformity issues.

Vaughn presented several real-world case studies demonstrating the challenges of PSD method transfer. As an example of instrument variation, he presented the particle size distribution of an API on two instruments, one which gave a D[v, 0.5] of 3.39 micron and the other which gave a D[v, 0.5] value of 3.62 micron. He also provided an example in differences between reporting mean, median, and mode averages and an example in which the same instrument from two different manufacturers resulted in a similar D[v, 0.5] values but differed significantly in their relative standard deviations (1.3% versus 9.7%). “The lessons learned are to assess the lab and the instrument, not the sample, and to always be aware of instrumental differences,” said Vaughn.

Manufacturers also must know the difference between transfer and site validation. “Although site validation offers added confidence, knowledge and ownership, and potentially stronger specifications, it also may lead to problems with local SOP issues because the CMO may have its own set of SOPs. On the other hand, method transfer is cheaper and faster and ‘relationship-friendly,’ but there is less knowledge and ownership,” said Vaughn.