Posted 04 December 2017 | By Zachary Brennan 

Evaluating Analytical Similarity: Stakeholders Raise Questions With FDA Draft Guidance

2700

Biopharma companies and industry groups, including Pfizer, Genentech, Boehringer Ingelheim and the Biosimilars Council, have raised detailed questions in recently released comments on the US Food and Drug Administration’s (FDA) draft guidance on statistical approaches to evaluate analytical similarities for biosimilars.

Back in September, FDA released the 15-page draft guidance on the type of information sponsors should obtain about the structural/physicochemical and functional attributes of a biosimilar’s reference product, how that information is used in the development of an analytical similarity assessment plan and the statistical approaches recommended for evaluating analytical similarity.

Martin Schiestl, chief science officer at Novartis' Sandoz, in October explained at a DIA conference how the US Food and Drug Administration's (FDA) draft guidance on statistical approaches to evaluate analytical similarity poses risks that could end up causing true biosimilars to be rejected randomly.

Comments

While acknowledging that the draft says there are many challenges and limitations to applying statistical analyses in the evaluation of analytical similarity data, Pfizer said in its comment that though the agency recommends a risk-based approach, "It is not clear how a risk-based approach in the analytical similarity assessment of quality attributes addresses the challenges outlined. Further, the challenge of determining an appropriate biologically or clinically meaningful margin for equivalence testing, which is fundamental to the utility of the outlined approach, is not addressed in the Draft Guidance."

Pfizer also noted, "The arbitrary nature of Tier 1 equivalence margin determination is a limitation for which a clear rationale is not provided in the Draft Guidance."

And the biopharma company also points to other challenges that are "not directly acknowledged or addressed" in the draft, including, "For example, the most substantial risk to the biosimilar developers’ ability to design the statistical analyses is the absence of control over the reference product, whose quality attribute levels may change at any time during the proposed biosimilar product’s development."

Pfizer also noted that throughout the draft, "The term anal

ytical similarity plan is used both for the statistical similarity plan and the application of analytical procedures. For clarity, these two concepts should be clearly distinguished throughout the text."

Boehringer Ingelheim said it agrees with FDA’s outlined approach, but the company laid out two principles that it believes should govern the use of analytic similarity assessment plans for all medicines.

"First, we believe it is imperative that FDA apply scientific and regulatory consistency to all biologics, including biosimilars and interchangeable biologics, to prevent any disparate treatment of these products. Second, FDA should continue to recognize that for all biologics prior knowledge (e.g., understanding of quality attributes of the biologic, molecular basis of the disease, understanding about vulnerable populations) is a critical element to consider for evaluation of similarity, and that this is particularly relevant for biosimilars and interchangeable biologics but not unique to them," BI’s comment said.

Roche’s Genentech, meanwhile, cited Focus’ story on Schiestl’s comments and called on FDA to consider limiting the scope of the draft to just non-orphan biosimilars, "removing all references to orphan drugs, and stressing the minimum sample sizes needed for the assessment. Another document could perhaps be written for orphan products."

From the perspective of industry groups, the Association for Accessible Medicines’ Biosimilars Council said the draft omits "what may be one of the most significant challenges and risks to the biosimilar developer. Specifically, the biosimilar developer has no control over the quality attributes of the reference product. Numerous studies have revealed changes over time in quality attributes of multiple reference products. It is important to acknowledge this risk in this paragraph as this can easily lead to shifts in the product mean that will impact equivalence testing."

The International Generic and Biosimilar Medicines Association also sought more flexibility in terms of alternative statistical or scientifically justified approaches.

Comments