Developers Ask FDA for Clarity on Lingering Concerns in Biosimilar Guidance

Fourteen parties have submitted comments on the guidance; in addition to other concerns with the document, including calls for clarification on the use of non-US comparator products and the potential for the waiver of bridging studies in the biosimilar development process, some drug makers called on the FDA to clarify issues associated with lot-to-lot variation and quality attributes.

In May 2018, the FDA published draft guidance related to comparative analytical assessment and other quality-related considerations for the development of therapeutic protein biosimilars. The document replaced previous draft guidance that was withdrawn in June 2018.

Fourteen parties have submitted comments on the guidance; in addition to other concerns with the document, including calls for clarification on the use of non-US comparator products and the potential for the waiver of bridging studies in the biosimilar development process, some drug makers called on the FDA to clarify issues associated with lot-to-lot variation and quality attributes.

In its comment letter, biosimilar developer Sandoz, together with its parent company, Novartis, said that it appreciates the FDA’s reconsideration of its statistical approach in favor of using quality ranges, calling it a practical approach that is consistent with regulations for biologic manufacturing.

However, Sandoz and Novartis called on the FDA to provide additional clarification. The letter notes that in cases in which products’ potency can vary among lots—in terms of antibody-dependent cell-mediated cytotoxicity or complement-dependent cytotoxicity, for example—different lots of reference standards may also vary in their relative potency. In the biosimilar development process, the letter states, “a change of the relative potency of representative product lots is especially common to the point where it is almost inevitable when the manufacturing process of such biosimilar candidates is modified in the course of development to resolve differences observed in the comparative analytical assessment.”

Because the guidance does not discuss how to assign the potency of reference standards that show lot-to-lot variation, the comment letter proposes to clarify that, for products that have variable potency, difference reference standards can be assigned by a stated potency that may not equal 100%. According to the drug makers, this proposal will be suitable to help control for drift.

Also raised in the Sandoz and Novartis comment letter is a concern that the draft guidance could hold biosimilars to stricter guidelines than those applied to reference products.

In the guidance’s discussion of the distribution of attributes, the agency says that it recommends that sponsors target the centers of distribution of the quality attributes of the reference product as closely as possible. The agency may be concerned about a distribution of an attribute that is biased to one side of the reference drug’s distribution, the guidance says, especially if the attribute has a role in the mechanism of action.

Depending on how the agency applies this requirement, says the Sandoz and Novartis letter, an “unintentional bias” among companies could arise. The letter proposes to allow the entire range of a given quality attribute for an FDA-approved reference product to represent acceptable quality. In an email to The Center for Biosimilars®, a Sandoz representative explained that “Market competition could be impacted if biosimilar makers have to hit a smaller target than the reference medicine does on an ongoing basis,” and said that “this may particularly impact more complex or orphan indications.”

However, said the representative, requirements can and do evolve; “We expect that in time, health authorities will become more familiar with biosimilar medicines, thereby requiring the same level of scrutiny.”