Critical Quality Attributes and non-CQAs are valuable to risk assessments, CASSS forum told.

BETHESDA, MD -- Risk assessments require both Critical Quality Attributes (CQAs) and non-CQAs to assure quality. At the July 24-25 CMC Strategy Forum held at the National Institutes of Health campus, representatives from biopharmaceutical companies and regulatory agencies discussed how criticality is determined and addressed.

This year, as manufacturers continue to evaluate the attributes they defined in 2007, it becomes increasingly clear that control work requires ample effort and resources. Whereas too much monitoring is not desirable, discounting or removing non-CQAs could affect the safety of a product in the long-term.

"When you talk about giving the CQAs to the agency only when you file the application ... I know you don't want to lock into things, because they will change," said Steve Kozlowski, M.D., director of the Office of Biotechnology Products at FDA's Center for Drugs. "But as we are all learning, if you plan on having CQAs, it would seem bringing them potentially into the [IND] discussions very early so the agency and the company are in accord on what they think is important and can be discussed, would be very useful." According to Milton Axley, Ph.D., associate director of process biochemistry, MedImmune: "Attributes help you define the target of what you want to make and design the process to make that product."

In the drug development and manufacturing industry, it is nearly impossible for scientists to know everything about their products. Researchers must consider non-CQAs in the process and product consistency realm. As low-risk criteria, non-CQAs are bound to draw less focus than CQAs, but there is no question that they are still a risk. Within a risk assessment, categorizing attributes to determine where they are needed the most will help manufacturers predict what could go wrong.

"Critical and non-critical experience can be important for us to approve a QBD," said Barbara Rellahan, Ph.D., principal investigator in the Division of Monoclonal Antibodies, Office of Biotechnology Products, CDER. According to FDA and biopharmaceutical experts, the best course of action when defining criticality is a classification system of low-, medium- and high-risk.

Distinguishing CQA is tricky, especially as the definition of "critical" can alternate between a broad concept and a highly subjective term. "If you take an antibody, there are very few things in the antibody that one could not call critical," said Siddharth Advant, Ph.D., head of the West Coast office at Tunnell Consulting. "One would be hard-pressed to find something that would not be critical. If one were to have a discussion of defining a CQA, it seems to be one that wouldn't have a very strong case or be able to test very easily."

Andrew Weiskopf, Ph.D., senior scientist of analytical development, Biogen Idec Inc., emphasized the importance of making connections between certain parameters. "If you establish a strong correlation between parameters that result in delivering a certain range of a particular attribute, and that attribute is considered a non-CQA, it can be controlled," he said.

The fact that scientists do not see CQAs as final, but as parameters that can be refined, expanded or reduced as manufacturing data is obtained, allows for some flexibility in risk assessments. According to Axley, drug developers are not stuck in their attribute definitions. "CQAs are not necessarily set at the beginning of the process; our understanding of attributes is going to evolve along the process," he said.

As CQA is a changing process, there are many factors manufacturers may deem critical at the beginning of a drug's lifecycle but reclassify as less critical down the line.

Carrie Nathans

Managing Editor