Challenges and approaches in demonstrating comparability of a well-characterized biotechnology product after manufacturing changes can be as varied and complex as the products themselves. Participants at the January 2005 CMC Strategy Forum sought to discuss and agree on common implementation strategies for different manufacturing change scenarios. Development of flexible, comprehensive approaches in strategy development addressed evaluation of critical product characteristics, appropriate process steps to test, numbers of lots and levels of testing required, and assessment of product comparability (e.g., trending analysis, additional characterization studies, accelerated stability data). The change scenarios we discussed can occur throughout the life-cycle of a product from early development through postapproval manufacturing.



PRODUCT FOCUS: WELL-CHARACTERIZED BIOLOGICS

PROCESS FOCUS: UPSTREAM AND DOWNSTREAM PROCESS DEVELOPMENT

WHO SHOULD READ: MANUFACTURING, PROCESS DEVELOPMENT, ANALYTICAL, AND PRODUCT DEVELOPMENT PERSONNEL

KEYWORDS: CHARACTERIZATION, CHANGE CONTROL, SPECIFICATIONS, DESIGN SPACE

LEVEL: INTERMEDIATE

Early stage development is where the foundation for assessing comparability begins, and the effects of good or poor development will carry throughout a product life-cycle. Sufficient process and product knowledge is required for reliably predicting and assessing the impact of a change and ensuring that a product will consistently meet approved specifications and standards. Efforts required to assess comparability are inversely proportional to a manufacturer's understanding of its manufacturing process, product quality attributes, and capability of the analytical methods used. An assessment of comparability should show that products are highly similar before and after a manufacturing change occurred. It does not mean the products are identical, but that their physicochemical and biological properties are sufficiently similar to ensure no adverse impact on their quality, safety, or efficacy.

Demonstrating Comparability



Forum Attendees Unanimously Agree: “Change Is Good!” The intention of this forum was to clarify for manufacturers the appropriate factors to consider when making changes to operations in pre- and postapproved manufacturing processes while ensuring patient safety, drug efficacy, and product quality. With all the issues to consider in demonstrating comparability, attendees unanimously agreed that despite associated hassles, the benefits of making changes far outweigh the costs of comparability studies.

Figure 1:

WWW.PHOTOS.COM

This familiar conundrum is reminiscent of a quote by famed cartoonist Sydney Harris: “Our dilemma is that we hate change and love it at the same time; what we want is for things to remain the same but get better.” As process knowledge and new technologies improve over time, manufacturers must follow. The needs of patients and caregivers also change over time and might require altering formulations or product presentations.

Regulatory agencies have agreed that it is reasonable for manufacturers to make changes in the steps of their manufacturing processes or specifications (testing and/or acceptance criteria) over the life-cycle of a biopharmaceutical product. Of the many benefits discussed at the forum, the main reasons for introducing process changes included process economics, product quality improvements, manufacturing technology advances, production yield or overall global capacity increases, and global harmonization of operating parameters. Improvements in test methods for process control and product characterization have driven evolution of the regulations. Innovation and improvements in manufacturing processes and test methods have been encouraged because they bring important and improved products to market efficiently and expeditiously. It was agreed that change is necessary for the biotech industry to evolve and progress, and international regulatory agencies see that as very positive outcome.

For well-characterized biotechnology products, comparability is the demonstration or existence of a high degree of similarity between products made by different manufacturing processes. Changes should have no adverse effects on the quality, safety, or efficacy of a product — the ultimate bottom line for demonstrating comparability. Several existing FDA, EU, and ICH guidance documents describe the principles of demonstrating comparability (1,2,3,4). A central tenet of those principles is that a given manufacturing change can be adequately assessed by comparing pre- and postchange materials and demonstrating that they are comparable (postchange material is of the same or better quality).

One forum attendee noted that for biologically derived products, better quality does not always mean greater purity. In certain products, impurities can act as stabilizers or enhance or inhibit activity. For example, attendees cited how a purer product may aggregate or cause an immunogenic response (as can a less-pure product).

Regulatory agencies encourage manufacturers to perform relatively extensive characterization in early phase development. The rationale is that such studies will be the primary focus in physicochemical characterization because not much information is known about clinical development and trials at that point. A thorough understanding of how manufacturing changes affect product quality attributes is desirable early in development because product–process relationships are important throughout the life-cycle of every product. Attendees acknowledged that obtaining early characterization data is resource dependent but often advantageous to manufacturers. Even if one product does not succeed to market, often a company can apply learning from it to similar modalities. (Monoclonal antibodies were cited as example.)

Changes in product or process may be more acceptable at early stages than after phase 3. Michael Klein of Amgen provided an excellent example of comparability assessment for a manufacturing change in early product development. The FDA's decision on comparability for the example was that the new product was sufficiently similar (though not comparable) to that used in earlier studies for its use to be allowed. Because the altered product was subject to further clinical evaluation, the potential impact of changes on safety and efficacy would be addressed. At later phases of development and postapproval, more stringent comparability criteria may be warranted.



PROCEEDINGS OF THE WCBP CMC STRATEGY FORUM, 9 JANUARY 2005

The fifth Well-Characterized Biotechnology Products (WCBP) Chemistry, Manufacturing, and Controls (CMC) Strategy Forum was held on 9 January 2005 at the Renaissance Mayflower Hotel in Washington, DC. Its purpose was discussion of issues related to demonstrating comparability for well-characterized biotechnology products in early and late development phases and postapproval. As with previous CMC Strategy Forums, the California Separation Sciences Society (CASSS; www.casss.org) sponsored this event. The organizers and moderators were John Towns (Eli Lilly) and Keith Webber (CDER, FDA). More than 130 attendees represented 40 large and small companies and 15 consulting firms as well as government agencies and academic organizations.

The forum consisted of a morning session devoted to the demonstration of comparability during preapproval and an afternoon session focused on postapproval comparability. Webber provided opening comments regarding the FDA's current regulatory guidelines and perspective related to demonstration of comparability. Michael Klein of Amgen provided an example of the evaluation of comparability preapproval by pointing to elements crucial to successful design and implementation of demonstrating comparability for product changes preapproval. Attendees then received three questions related to preapproval comparability that facilitated discussion for the remainder of the session.

The afternoon session began with two case studies by Tina Norsell of Novo Nordisk on the investigation of impurities in a biopharmaceutical product. Genentech's John O'Conner followed with a presentation on demonstrating comparability for products manufactured at different global facilities. Allison Wolf of Eli Lilly provided a case study for successfully reducing a product's reporting category by using a comparability protocol to reduce the reporting category down from a prior-approval supplement (PAS) to an annual reportable (AR) change. Attendees then received three questions relating to postapproval comparability that facilitated discussion for the remainder of this session.

Bioassays and/or pharmacokinetic and pharmacodynamic studies may be required for biotechnology-derived products because many physicochemical tests cannot accurately detect small changes in such products. Animal testing may be required if in vitro bioassays do not detect potential changes in a product's tertiary structure. If physicochemical comparability cannot be demonstrated with a production batch, then the change has substantial potential to adversely affect the identity, strength, quality, purity, and/or potency of a drug product. An applicant still wishing to institute such a change would perform additional studies to assess the impact of that change. Those might include in vitro or in vivo biological studies or even human clinical trials.

Attendees emphasized that postchange statistical impurity limits should be within the range experienced in clinical trials. When impurities reach levels beyond what was observed clinically, the decision most likely to be made is to alter a process further and get those impurities back within the range used for clinical trials. Also, with increased manufacturing experience, the variance of product characteristics often decreases. Manufacturers may consider narrowed ranges as a benchmark during comparability assessment if that helps them ensure consistent product quality.

Implementing substantial manufacturing process changes after phase 3 is considered a special case that generally requires submission of data to demonstrate comparability in a marketing application. The FDA cited such changes as a special case because they come after completion of product safety and efficacy studies. The agency realizes that it is not always possible to delay implementing a change, but doing so increases the amount of information that must be reviewed prior to approval.

Attendees said that major failures in demonstrating comparability for a change were often caused by inadequate training of new people as well as subtle differences in new equipment and raw materials. Under the training of personnel, a subtopic of transferring analytical methods was cited as a large cause of variability when making a manufacturing site change. In extreme cases methods could not be transferred, so originating laboratories have been stuck running a new site's tests. Even hiccups extrinsic to a process (e.g., a faulty flow valve with a column resin change) need to be investigated and resolved when the flow of consecutive lots is interrupted. Another example cited occurred when the composition of media needed to be adjusted during demonstration lots to obtain a similar glycosylation pattern (in this case galactose levels) to that observed before a manufacturing change. Lower galactose levels were not a problem of the expression system but came from stress the cells were under as the manufacturer worked toward higher titers. Higher galactose levels were easily achieved by increasing certain nutrient levels in the media.

Figure 1 shows the knowledge base and testing curves as a function of a product life-cycle. The knowledge base increases as a manufacturer learns more about critical process parameters and product quality attributes. Meanwhile, overall testing (physicochemical, biological, toxicological, and clinical) decreases. Early development involves a full arsenal of testing from animal tests to clinical trials and physicochemical and/or biological testing. As the knowledge base increases, the testing curve decreases. In assessments of comparability for postapproval manufacturing changes, the testing requirements often can be limited to physical, chemical, and biological testing.